Journal of Speech and HearingResearch, Volume 33, 149-155, March 1990
HEARING LOSS, AGING, AND SPEECH PERCEPTION IN REVERBERATION AND NOISE KAREN S. HELFER University of Massachusetts at Amherst
LAURA A. WILBER Northwestern University
The present investigation examined the effect of reverberation and noise on the perception of nonsense syllables by four groups of subjects: younger (35 years of age) and older (>60 years of age) listeners with mild-to-moderate sensorineural hearing loss; younger, normal-hearing individuals; and older adults with minimal peripheral hearing loss. Copies of the Nonsense Syllable Test (Resnick, Dubno, Huffnung, & Levitt, 1975) were re-recorded under four levels of reverberation (0.0, 0.6, 0.9, 1.3 s) in quiet and in cafeteria noise at + 10 dB S:N. Results suggest that both age and amount of pure-tone hearing loss contribute to senescent changes in the ability to understand noisy, reverberant speech: pure-tone threshold and age were correlated negatively with performance in reverberation plus noise, although age and pure-tone hearing loss were not correlated with each other. Further, many older adults with minimal amounts of peripheral hearing loss demonstrated difficulty understanding distorted consonants. KEY WORDS: aging, speech perception, reverberation
Speech is seldom transmitted in a totally quiet, echofree environment. Most listening situations have some noise and some degree of reverberation. Both of these distortions greatly alter a speech signal, although they produce somewhat different effects: noise obscures the less intense portions of a stimulus, whereas reverberation causes masking of adjacent phonemes, smears elements in the time domain, and smooths the temporal envelope (e.g., Houtgast & Steeneken, 1973). Although young normal-hearing adults can tolerate moderate amounts of noise (Cooper & Cutts, 1970; Olsen, Noffsinger, & Kurdziel, 1975) and reverberation (Crum, 1974; Nabelek & Jennings, 1981; Nabelek & Pickett, 1974) with only minimal degradation of their speech processing abilities, individuals with sensorineural hearing loss appear to be much more susceptible to these distortions (Hawkins & Yacullo, 1984; Humes, Dirks, Bell, Ahlstrom, & Kincaid, 1987; Nabelek & Dagenais, 1986; Nabelek & Pickett, 1974; Nabelek & Robinette, 1978a, 1978b; Palva, 1955; Ross, Huntington, Newby, & Dixon, 1965). This is especially apparent for older individuals with presbycusis (Bergman, 1980; Duquesnoy, 1983; Duquesnoy & Plomp, 1980; Harris & Reitz, 1985; Nabelek & Robinson, 1982; Plomp & Duquesnoy, 1980; Plomp & Mimpen, 1979; Smith & Prather, 1971). When noise and reverberation are combined (as occurs frequently in actual listening situations) even younger listeners with normal auditory systems experience difficulty with speech understanding (Irwin & McCauley, 1987; Moncur & Dirks, 1967; Nabelek & Pickett, 1974; Nabelek & Robinette, 1978a, 1978b). Age-related decline in the absence of substantial puretone hearing loss has been noted on measures of consonant perception in quiet (Gelfand, Ross, & Miller, 1986) and in noise (Dubno, Dirks, & Morgan, 1984; Gelfand et al., 1986). Limited data indicate that such individuals also experience difficulty perceiving speech sounds distorted by reverberation (Nabelek, 1988; Nabelek & Robinson, 1982) and by reverberation plus noise (Harris & Reitz, © 1990, American Speech-Language-Hearing Association
1985). These results suggest that older individuals with little pure-tone hearing loss may be experiencing difficulty in real rooms. Because presbycusis can affect any auditory system level (Schuknecht, 1955) senescent changes in speech perception without accompanying pure-tone deficits may be attributed to central auditory decline, but also may be due to peripherally mediated psychoacoustic changes (e.g., frequency resolution and temporal resolution) not measured by the pure-tone audiogram. Several researchers have noted marked variability on tasks of speech perception in reverberation, particularly among hearing-impaired (Nabelek & Letowski, 1985; Nabelek & Pickett, 1974) and older (Bergman, 1980) listeners. Susceptibility to reverberation is reported to be unrelated to degree of hearing loss (Nabelek & Robinette, 1978a). In addition, investigations suggest that performance in noisy, reverberant conditions cannot be predicted from performance in noise alone (Nabelek & Mason, 1981) or from scores obtained using simulated reverberation (Irwin & McCauley, 1987; Nabelek & Robinette, 1978a). Thus, on the basis of data obtained during a typical audiologic evaluation, the ability to predict how successfully a given individual will function outside of the test suite is limited. Psychoacoustic abilities and cognitive/ memory skills not measured during audiologic evaluation are likely to play a role in deciphering a distorted message. For example, results of a recent study (Irwin & McCauley, 1987) support the link between temporal resolution and the recognition of reverberated speech by hearing-impaired and normal-hearing adults. At present, it is not known whether an older person's perceptual difficulty in reverberation/noise is caused by the aging process or by the distortion that accompanies all sensorineural hearing loss. The present investigation represents an attempt to examine the influences of both age and pure-tone hearing loss on the perception of nonsense syllables distorted by reverberation and noise. 149
Downloaded From: http://jslhr.pubs.asha.org/ by a Western Michigan University User on 03/27/2016 Terms of Use: http://pubs.asha.org/ss/rights_and_permissions.aspx
0022-4685/90/3301-0 149$01.00/0
33
150 Journal of Speech and Hearing Research METHODS
Generation of Speech Stimuli Because older subjects might differ from each other (and from the younger participants) in linguistic background, intellect, memory, and response bias, a closed-set nonsense syllable test was employed in an attempt to control the influence of these variables. Although the use of such a task limits direct application to real-life performance (listeners cannot call upon contextual cues or their linguistic knowledge to aid in perception of nonsense syllables) it does allow for documentation of senescent changes on a basic phonemic level. The specific speech perception task selected for this investigation was the Nonsense Syllable Test (NST) developed by Resnick et al. (1975). The NST consists of 91 syllables divided into 11 subsets; each subset contains between 7 and 9 syllables. The subsets differ from each other in class of consonant (voiced or voiceless), position of consonant (VC or CV), and vowel (/a/, /i/, or /u/). The NST stimuli are embedded within the carrier phrase, please." "You will mark A copy of the NST was digitized (using a 20-kHz sampling rate) and randomized into eight complete versions. After being converted back to analog form, the randomized NST lists were re-recorded in a room having variable reverberation time. The reverberation room used for this purpose was a steel-walled enclosure of 6000 cubic feet (20 x 25 x 12 feet) designed to allow for controllable amounts of reverberation by attaching panels of absorptive material to the exposed surfaces of the room. To ensure that the stimuli consisted of primarily reverberant energy, all reverberation measurements and subsequent recordings were completed using a 12-foot loudspeaker-to-microphone distance (the critical distance in this room was between 6 feet and 12 feet). Reverberation was quantified using two different procedures. Manual reverberation measurements were completed by introducing pure tones and narrow bands of noise into the enclosure and measuring the decay pattern displayed on a graphic level recorder. Reverberation time (the time necessary for sound in the room to decay by 60 dB) was extrapolated from the 20-dB or 30-dB decay point. Reverberation was also measured using the Crown Tecron TEF System 10 Audio Spectrum Analyzer. The Energy Time Curve program of this computer-aided system measures reflections in a given frequency band, performs a Fast Fourier Transform on the data, then fits a regression line to the decay pattern to yield a reverberation time value. The stimuli used by the TEF system were sine waves swept over 200-Hz-wide bands centered on octave frequencies between 125 Hz and 8000 Hz. The agreement between reverberation time (RT) values obtained using the two methods decreased as RT increased. Therefore, nominal RT values were calculated by averaging the average RT figures at 500, 1000, and
149--155
March 1990
2000 Hz derived from the two measurement procedures. Nominal RT values used for generating recorded material were 0.6 s for the least reverberant condition, 0.9 s for the moderate reverberation condition, and 1.3 s for the most reverberant condition, representative of RT in small, medium, and large rooms (Lochner & Burger, 1961). One entire NST set was re-recorded twice at each of the three reverberation times: once in quiet, and once in the presence of the cafeteria competition (at +10 dB S:N relative to the word "mark" in the carrier phrase) recorded on the second track of the original NST tapes. Resnick et al. (1975) describe this noise as having a relatively flat power spectrum up to 600 Hz, beyond which energy declines at the rate of 10 dB per octave (see Resnick et al., 1975 for a complete description of this noise). The NST stimuli were played out of a cassette recorder (Sony TC-D5M), amplified (NAD Series 20 Stereo Amplifier), then fed to a loudspeaker (MDM TA-2) placed in the reverberation room at approximately 0° re: a Knowles Electronics Mannequin for Acoustic Research (KEMAR). The exact positioning of KEMAR was determined by measuring the place where the sound pressure level (SPL) outside of KEMAR's two ear canals was the same (this was necessary because of the non-uniform reflection pattern in the reverberation chamber). Two loudspeakers, placed at approximately 450 left and 45 ° right of KEMAR at a distance of 14 feet, transmitted the competition. Stimuli were picked up by microphones (Etymotic Research ER-11) placed in the ear canals of KEMAR, sent through the associated preamplifier designed to remove ear canal resonances (Etymotic Research ER-11), then recorded digitally onto videotape using a Pulse-Code Modulation (Sony PCM 501 ES/NEC VCR VC-N40 EU) recorder. In an attempt to preserve binaural cues the output of KEMAR's right ear was sent to track 1 of the PCM; the left-ear output was routed to track 2. The entire taping procedure was repeated in an anechoic chamber to obtain two lists (one in quiet and one with competition) having essentially no reverberation. Procedures Each subject listened to all eight of the NST lists (0.0, 0.6, 0.9, 1.3 s RT in quiet, and at +10 dB S:N). The NST stimuli were played out of the PCMNCR unit through a clinical audiometer (Grason-Stadler 1701) and were delivered to the listener via TDH-49 headphones. Track 1 of the test tape (with the output from KEMAR's right ear) was sent to the right headphone, and Track 2 of the tape (containing the output from KEMAR's left ear) was routed to the left headphone. The stimuli were delivered at 50 dB re: the pure-tone average, or at the highest level judged comfortable by the participant. Subjects responded by circling their choices on a response form. No feedback was employed, and subjects were instructed to circle a response for each stimulus. Presentation order was randomized across the eight conditions for each subject.
Downloaded From: http://jslhr.pubs.asha.org/ by a Western Michigan University User on 03/27/2016 Terms of Use: http://pubs.asha.org/ss/rights_and_permissions.aspx
HELFER & WILBER: Reverberation and Noise
Subjects Participants were younger (less than 36 years of age) and older (over 60 years of age) ambulatory adults with a negative history of neurological disorder, middle ear pathology, and ototoxic drug usage. In an attempt to obtain a sample of individuals with presbycusis, older subjects were also screened for excessive noise exposure. In addition, none of the subjects demonstrated an asymmetric hearing loss, and all participants learned English as their first language. Participants were divided into four groups, with 8 subjects per group. Table 1 contains age and pure-tone average data for the subject groups. The YN (young normal) group consisted of young adults with no evidence of auditory system disorder. Participants in this group met the following criteria: pure-tone thresholds s20 dB HL (ANSI, 1969) for test frequencies between 250 Hz and 8000 Hz; air-bone gaps .01) for the reverberation plus noise condition for all groups. For reverberation alone, the difference in scores between the two highest reverberation levels was significant only for the OHI group. Thus, although the 1.3 RT recordings were subjectively more distorted than the 0.9 RT lists, the decline in performance from reverberation appeared to plateau at 0.9 RT. A plateau effect in reverberation is also apparent in the data from Nabelek and Robinson (1982) for reverberation alone. Comparison of performance between the OHI and YHI groups is confounded by the difference in amount of hearing loss between these two sets of subjects; discerning aging effects is difficult because of the greater amount of hearing loss in the YHI group. However, other group comparisons were completed using Scheffe post-hoc tests. Performance by the YN group was significantly (p < .01) better than that of the other three subject groups for all NST conditions. Scores obtained by the two groups of older subjects differed significantly (p < .01) for the 0.OQ, 0.ON, 0.9N, and 1.3N conditions; participants in the OM group obtained similar scores to the older adults with greater amounts of hearing loss for stimuli distorted by reverberation alone and mild amounts of reverberation plus noise.
Downloaded From: http://jslhr.pubs.asha.org/ by a Western Michigan University User on 03/27/2016 Terms of Use: http://pubs.asha.org/ss/rights_and_permissions.aspx
153
HELFER & WILBER: Reverberation and Noise
where RScore also showed an inverse relationship with age. Thus, subject age appeared to be an important variable in distorted NST perception, although age was not correlated with hearing loss or with undistorted nonsense syllable perception. Amount of hearing loss was also an important contributor to performance in distortion: both PTA1 and PTA2 were correlated negatively with the average reverberation and reverberation plus noise scores. PTA1 and PTA2 were also correlated highly with each other. NST performance in quiet with no reverberation was correlated positively with MaxRN, RScore, RNScore, and with NST in noise, and correlated negatively with both measures of pure-tone average. When threshold was partialed out the undistorted NST score remained correlated with RNScore, MaxRN, and with NST in noise, although the correlation coefficients were lower. Thus, percent-correct performance on this reverberated nonsense syllable task was related, on an individual basis, to performance on the same task in its undistorted condition. It is not known whether specific consonant error patterns (not analyzed in the present study) were similar for reverberated and unreverberated stimuli. NST distorted by noise alone was correlated negatively with age, PTA1 and PTA2, and correlated positively with undistorted NST, RScore, and RNScore. When PTA1 and PTA2 were partialed out these correlations continued to be significant. The partial correlation matrix also contained a negative correlation between performance in noise alone and percent decline from noise (MaxN).
The percent-decline indices and averaged scores calculated from the NST data were subjected to separate ANOVAs. Highly significant group effects were noted for both RScore [F(3, 28) = 15.745; p < .0001] and for RNScore [F(3, 28) = 21.976; p = .0001]. Post-hoc Scheffe comparisons revealed the YN group to have significantly higher RScore and RNScore values (p < .01), as compared to the other three groups. Group differences in MaxR, MaxN, and MaxRN failed to reach statistical significance; the amount of decline in consonant perception caused by reverberation and noise did not vary systematically among subject groups. CorrelationAnalyses Pearson r and partial correlations were calculated to examine interrelations among the variables. The partial correlations were computed with the two measures of pure-tone average controlled. The correlation matrix is displayed in Table 2. The correlation analyses used the following variables: age, PTA1, PTA2, RScore, RNScore, MaxR, MaxN, MaxRN, NST in quiet with no reverberation (0.0Q), and NST in noise alone (0.ON). The individual NST scores in reverberation and in reverberation plus noise were not utilized for two reasons: first, significant correlations would probably have been obscured because of a relatively large number of variables as compared to the number of subjects; second, the scores on the individual NST lists would have been intercorrelated. Therefore, the average indices (RScore and RNScore) were utilized for all subsequent analyses. The correlation results showed age to be correlated negatively with RNScore and with NST in noise. These correlations persisted in the partial correlation analysis,
DISCUSSION
AND CONCLUSIONS
The majority of older participants in this study, including several individuals with only mild peripheral hearing
TABLE 2. Correlation results for data pooled across all subjects. The upper value in the matrix is the Pearson product-moment coefficient;
the lower value is the correlation coefficient obtained when PTA1 and PTA2 are partialed out (* = statistical significance at the .01 level; **= statistical significance at the .001 level). Age Age
--
NST/Quiet
NST/Quiet -. 292 -. 371 -
-RScore RNScore
RScore
RNScore
NSTINoise
MaxR
MaxN
MaxRN
-. 433* -. 676**
-. 453* -. 631**
.112 .242
.252 .307
.077 .150
-. 025 -
.885**
.869**
.865**
-. 067
.395
.435*
-. 749**
-. 875**
.560** -
.469* .920**
.525** .869**
.392 -. 492*
.423* .159
.469* .135
-. 879**
-. 879**
--
.725**
.609**
-. 460*
-. 099
-. 161
.954**
-. 388
- .026
-.004
.851**
-. 332
-. 445*
-. 438*
-. 304
-. 120
-. 176
-. 549**
-
.423*
.549**
.572**
.689** .690** .675**
-. 255 -. 517*
--
NST/Noise MaxR MaxN
--
--
MaxRN PTA1
PTA2
Downloaded From: http://jslhr.pubs.asha.org/ by a Western Michigan University User on 03/27/2016 Terms of Use: http://pubs.asha.org/ss/rights_and_permissions.aspx
.093 -. 202
PTA1
PTA2 .148
-
-. 758**
-. 861**
-
-. 707** -
.452* -. 188 -.161 -
-. 823** -
.247 -. 223 -. 250 .837**
154 Journal of Speech and HearingResearch
loss, demonstrated marked difficulty understanding nonsense syllables distorted by reverberation and noise. Perhaps the most revealing results was the strong negative correlation between age and performance in reverberation and noise. A connection between age and speech perception in reverberation and noise was also noted by Harris and Reitz (1985) and by Nabelek (1988), especially for conditions with severe amounts of distortion. In fact, in the reverberation plus noise conditions performance by the older hearing-impaired participants was slightly less accurate than that of the younger hearing-impaired listeners, even though pure-tone threshold was correlated with NST scores in distortion and the younger subjects had significantly greater hearing loss. Thus, although peripheral hearing loss was an important contributor to the decline noted in distorted conditions, some factor(s) other than amount of threshold elevation also appeared to be limiting the performance of these older listeners. The reverberation and noise levels employed in this investigation were typical of those encountered in real rooms. This amount of distortion was sufficient to interfere with older adults' ability to perceive consonants. Certainly, the deciphering on nonsense syllables is a difficult, somewhat artificial task, because the listener cannot use linguistic knowledge to fill in distorted or missing portions of the message. At present, it is unknown whether older adults also demonstrate difficulty processing realistic stimuli (i.e., running speech) distorted by reverberation plus noise. The deficit in perception noted in the performance of some of the older adults with minimal hearing loss confirms previous reports of age-related decline in intelligibility of distorted speech. As amount of distortion increases, elderly individuals with little peripheral hearing loss appear to perform less like young, normalhearing listeners (Harris & Reitz, 1985; Nabelek, 1988; Nabelek & Robinson, 1982). Variability, however, is a hallmark in the aging literature, and was observed in our sample of older subjects (several attained scores similar to those obtained by young, normal hearing listeners). Additional research should focus on identifying auditory and non-auditory abilities contributing to age-related speech perception deficits. Correlational analyses revealed a strong connection between pure-tone hearing loss and consonant perception in distortion. A significant relation also existed between both performance in quiet or in noise and NST perception in reverberation plus noise. Because a clinical measure of speech perception in reverberation/noise is not yet available, it would be advantageous to be able to predict performance under realistic listening conditions from data obtained from conventional audiologic tests. Due to the relatively large amount of intersubject variability noted in this and other studies of speech perception in reverberation, it would be unwise to assume that all individuals who perform normally on typical audiologic measures have little difficulty perceiving speech distorted by room acoustics. Although, in this investigation, pure-tone hearing loss and performance on a mea-
33 149-155
March 1990
sure of speech in its undistorted state were related to scores obtained in reverberation/noise, future studies should examine the relation between performance on reverberated and unreverberated versions of other speech material.
ACKNOWLEDGMENTS This article is based upon a doctoral dissertation completed by the first author under the direction of the second author, and was supported in part by a Dissertation Year Fellowship from Northwestern University. We wish to thank Bill Martens, Dean Garstecki, Tom Carrell, and Bill Yost for their assistance on this project and Harry Levitt for supplying the NST tapes. We also thank Rich Freyman, Anna Nabelek, and Peter Fitzgibbons for their comments on an earlier version of the manuscript. Portions of this paper were presented at the American Speech-LanguageHearing Association convention in Boston, Massachusetts, November, 1988. REFERENCES AMERICAN NATIONAL STANDARDS INSTITUTE. (1969). Specificationsfor audiometers (S3.6-1969). New York: ANSI. BERGMAN, M. (1980). Aging and the perception of speech. Baltimore: University Park Press. COOPER, J. C., & CUTTS, B. P. (1970). Speech discrimination in noise. Journal of Speech and HearingResearch, 13, 332-337. CRUM, M. (1974). Effects of reverberation, noise and distance upon speech intelligibility in small, classroom size acoustic enclosures. Unpublished doctoral dissertation, Northwestern University, Evanston, IL. DUBNO, J. R., DIRKS, D. D., & MORGAN, D. E. (1984). Effects of age and mild hearing loss on speech recognition in noise. Journal of the Acoustical Society of America, 76, 87-96. DUQUESNOY, A. J. (1983). The intelligibility of sentences in quiet and in noise in aged listeners. Journalof the Acoustical Society of America, 74, 1136-1144. DUQUESNOY, A. J., & PLOMP, R. (1980). Effect of reverberation and noise on the intelligibility of sentences in cases of presbyacusis. Journal of the Acoustical Society of America, 68, 537-544. GELFAND, S. A., Ross, L., & MILLER, S. (1986). Effects of aging and hearing loss on sentence reception in noise from one or two sources. Journal of the Acoustical of America, 80, s77. HARRIS, R. W., & REITZ, M. L. (1985). Effects of room reverberation and noise on speech discrimination by the elderly. Audiology, 24, 319-324. HAWKINS, D. B., & YACULLO, W. S. (1984). Signal-to-noise ratio advantage of binaural hearing aids and directional microphones under different levels of reverberation. Journal of Speech and Hearing Disorders,49, 278-286. HOUTGAST, T., & STEENEKEN, H. J. M. (1973). The modulation transfer function in room acoustics as a predictor of speech intelligibility. Acustica, 28, 66-73. HUMES, L. E., DIRKS, D. D., BELL, T. S., AHLSTROM, C., & KINCAID, G. E. (1986). Application of the Articulation Index and the Speech Transmission Index to the recognition of speech by normal-hearing and hearing-impaired listeners. Journal of Speech and HearingResearch, 29, 447-462. IRWIN, R. J., & MCCAULEY, S. F. (1987). Relations among temporal acuity, hearing loss, and the perception of speech distorted by noise and reverberation. Journal of the Acoustical Society of America, 81, 1557-1565. ISO (1982). Acoustics-Thresholdof hearingby airconduction as a function of age and sex for otologically normal persons. ISO/DIS 7029. LOCHNER, J. P. A., & BURGER, J. F. (1961). The intelligibility of
Downloaded From: http://jslhr.pubs.asha.org/ by a Western Michigan University User on 03/27/2016 Terms of Use: http://pubs.asha.org/ss/rights_and_permissions.aspx
HELFER & WILBER: Reverberation and Noise speech under reverberant conditions. Acustica, 11, 195-200. MONCUR, J. P., & DIRKS, D. (1967). Binaural and monaural speech intelligibility in reverberation. Journal of Speech and HearingResearch, 10, 186-195. NABELEK, A. K. (1988). Identification of vowels in quiet, noise, and reverberation: Relationships with age and hearing loss. Journalof the Acoustical Society of America, 84, 476-484. NABELEK, A. K., & DAGENAIS, P. A. (1986). Vowel errors in noise
and in reverberation by hearing-impaired listeners. Journal of the Acoustical Society of America, 80, 741-748. NABELEK, A. K., & JENNINGS, E. (1981). Age effects on speech
perception in reverberation. Journal of the Acoustical Society of America, 72, s89. NABELEK, A. K., & LETOWSKI, T. R. (1985). Vowel confusions of hearing-impaired listeners under reverberant and non-reverberant conditions. Journal of Speech and Hearing Disorders, 50, 126-131. NABELEK, A. K., & MASON, D. (1981). Effect of noise and
reverberation on binaural and monaural word identification by subjects with various audiograms. Journal of Speech and Hearing Research, 24, 375-383. NABELEK, A. K., &PICKETT, J. M. (1974). Monaural and binaural
speech perception through hearing aids under noise and reverberation with normal and hearing-impaired listeners. Journalof Speech and HearingResearch, 17, 724-739. NABELEK, A. K., & ROBINETTE, L. (1978a). Reverberation as a
parameter in clinical testing. Audiology, 17, 239-259. NABELEK, A. K., & ROBINETTE, L. (1978b). Influence of the
155
OLSEN, W. O., NOFFSINGER, D., & KURDZIEL, S. (1975). Speech
discrimination in quiet and in white noise by patients with peripheral and central lesions. Acta Otolaryngologica, 80, 375-382. PALVA, A. (1955). Studies of hearing for pure-tones and speech in noise. Acta Otolaryngologica,70, 232-241. PLOMP, R., & DUQUESNOY, A. J. (1980). Room acoustics for the aged. Journal of the Acoustical Society of America, 68, 16161621. PLOMP, R., & MIMPEN, A. M. (1979). Speech-reception threshold for sentences as a function of age and noise level. Journalof the Acoustical Society of America, 66, 1333-1342. RESNICK, J. B., DUBNO, J. R., HOFFNUNG, S., & LEVITT, H.
(1975). Phoneme errors on a nonsense syllable test. Journalof the Acoustical Society of America, 58, (Suppl. 1), 114. Ross, M., HUNTINGTON, D. A., NEWBY, H. A., & DIXOQN, R. F.
(1965). Speech discrimination of hearing-impaired individuals in noise. Journal of Auditory Research, 5, 47-72. SCHUKNECHT, H. (1955). Presbycusis. Laryngoscope, 65, 407-
419. SILMAN, S., & GELFAND, S. A. (1981). The relationship between magnitude of hearing loss and acoustic reflex threshold levels. Journal of Speech and Hearing Disorders, 46, 312-316. SMITH, R. A., &PRATHER, W. F. (1971). Phoneme discrimination
in older persons under varying signal-to-noise conditions. Journal of Speech and Hearing Research, 14, 630-638. WINER, B. J. (1971). Statisticalprinciplesin experimentaldesign (2nd ed.). New York: McGraw-Hill.
precedence effect on word identification by normally hearing and hearing-impaired subjects. Journal of the Acoustical Society of America, 63, 187-194.
Received March 6, 1989 Accepted August 7, 1989
NABELEK, A. K., & ROBINSON, P. K. (1982). Monaural and binau-
ral speech perception in reverberation for listeners of various ages. Journal of the Acoustical Society of America, 71, 12421248.
Request for reprints should be sent to Karen S. Helfer, University of Massachusetts, Department of Communication Disorders, Arnold House, Amherst, MA 01003.
Downloaded From: http://jslhr.pubs.asha.org/ by a Western Michigan University User on 03/27/2016 Terms of Use: http://pubs.asha.org/ss/rights_and_permissions.aspx
HEARING LOSS, AGING, AND SPEECH PERCEPTION IN REVERBERATION AND NOISE KAREN S. HELFER University of Massachusetts at Amherst
LAURA A. WILBER Northwestern University
The present investigation examined the effect of reverberation and noise on the perception of nonsense syllables by four groups of subjects: younger (35 years of age) and older (>60 years of age) listeners with mild-to-moderate sensorineural hearing loss; younger, normal-hearing individuals; and older adults with minimal peripheral hearing loss. Copies of the Nonsense Syllable Test (Resnick, Dubno, Huffnung, & Levitt, 1975) were re-recorded under four levels of reverberation (0.0, 0.6, 0.9, 1.3 s) in quiet and in cafeteria noise at + 10 dB S:N. Results suggest that both age and amount of pure-tone hearing loss contribute to senescent changes in the ability to understand noisy, reverberant speech: pure-tone threshold and age were correlated negatively with performance in reverberation plus noise, although age and pure-tone hearing loss were not correlated with each other. Further, many older adults with minimal amounts of peripheral hearing loss demonstrated difficulty understanding distorted consonants. KEY WORDS: aging, speech perception, reverberation
Speech is seldom transmitted in a totally quiet, echofree environment. Most listening situations have some noise and some degree of reverberation. Both of these distortions greatly alter a speech signal, although they produce somewhat different effects: noise obscures the less intense portions of a stimulus, whereas reverberation causes masking of adjacent phonemes, smears elements in the time domain, and smooths the temporal envelope (e.g., Houtgast & Steeneken, 1973). Although young normal-hearing adults can tolerate moderate amounts of noise (Cooper & Cutts, 1970; Olsen, Noffsinger, & Kurdziel, 1975) and reverberation (Crum, 1974; Nabelek & Jennings, 1981; Nabelek & Pickett, 1974) with only minimal degradation of their speech processing abilities, individuals with sensorineural hearing loss appear to be much more susceptible to these distortions (Hawkins & Yacullo, 1984; Humes, Dirks, Bell, Ahlstrom, & Kincaid, 1987; Nabelek & Dagenais, 1986; Nabelek & Pickett, 1974; Nabelek & Robinette, 1978a, 1978b; Palva, 1955; Ross, Huntington, Newby, & Dixon, 1965). This is especially apparent for older individuals with presbycusis (Bergman, 1980; Duquesnoy, 1983; Duquesnoy & Plomp, 1980; Harris & Reitz, 1985; Nabelek & Robinson, 1982; Plomp & Duquesnoy, 1980; Plomp & Mimpen, 1979; Smith & Prather, 1971). When noise and reverberation are combined (as occurs frequently in actual listening situations) even younger listeners with normal auditory systems experience difficulty with speech understanding (Irwin & McCauley, 1987; Moncur & Dirks, 1967; Nabelek & Pickett, 1974; Nabelek & Robinette, 1978a, 1978b). Age-related decline in the absence of substantial puretone hearing loss has been noted on measures of consonant perception in quiet (Gelfand, Ross, & Miller, 1986) and in noise (Dubno, Dirks, & Morgan, 1984; Gelfand et al., 1986). Limited data indicate that such individuals also experience difficulty perceiving speech sounds distorted by reverberation (Nabelek, 1988; Nabelek & Robinson, 1982) and by reverberation plus noise (Harris & Reitz, © 1990, American Speech-Language-Hearing Association
1985). These results suggest that older individuals with little pure-tone hearing loss may be experiencing difficulty in real rooms. Because presbycusis can affect any auditory system level (Schuknecht, 1955) senescent changes in speech perception without accompanying pure-tone deficits may be attributed to central auditory decline, but also may be due to peripherally mediated psychoacoustic changes (e.g., frequency resolution and temporal resolution) not measured by the pure-tone audiogram. Several researchers have noted marked variability on tasks of speech perception in reverberation, particularly among hearing-impaired (Nabelek & Letowski, 1985; Nabelek & Pickett, 1974) and older (Bergman, 1980) listeners. Susceptibility to reverberation is reported to be unrelated to degree of hearing loss (Nabelek & Robinette, 1978a). In addition, investigations suggest that performance in noisy, reverberant conditions cannot be predicted from performance in noise alone (Nabelek & Mason, 1981) or from scores obtained using simulated reverberation (Irwin & McCauley, 1987; Nabelek & Robinette, 1978a). Thus, on the basis of data obtained during a typical audiologic evaluation, the ability to predict how successfully a given individual will function outside of the test suite is limited. Psychoacoustic abilities and cognitive/ memory skills not measured during audiologic evaluation are likely to play a role in deciphering a distorted message. For example, results of a recent study (Irwin & McCauley, 1987) support the link between temporal resolution and the recognition of reverberated speech by hearing-impaired and normal-hearing adults. At present, it is not known whether an older person's perceptual difficulty in reverberation/noise is caused by the aging process or by the distortion that accompanies all sensorineural hearing loss. The present investigation represents an attempt to examine the influences of both age and pure-tone hearing loss on the perception of nonsense syllables distorted by reverberation and noise. 149
Downloaded From: http://jslhr.pubs.asha.org/ by a Western Michigan University User on 03/27/2016 Terms of Use: http://pubs.asha.org/ss/rights_and_permissions.aspx
0022-4685/90/3301-0 149$01.00/0
33
150 Journal of Speech and Hearing Research METHODS
Generation of Speech Stimuli Because older subjects might differ from each other (and from the younger participants) in linguistic background, intellect, memory, and response bias, a closed-set nonsense syllable test was employed in an attempt to control the influence of these variables. Although the use of such a task limits direct application to real-life performance (listeners cannot call upon contextual cues or their linguistic knowledge to aid in perception of nonsense syllables) it does allow for documentation of senescent changes on a basic phonemic level. The specific speech perception task selected for this investigation was the Nonsense Syllable Test (NST) developed by Resnick et al. (1975). The NST consists of 91 syllables divided into 11 subsets; each subset contains between 7 and 9 syllables. The subsets differ from each other in class of consonant (voiced or voiceless), position of consonant (VC or CV), and vowel (/a/, /i/, or /u/). The NST stimuli are embedded within the carrier phrase, please." "You will mark A copy of the NST was digitized (using a 20-kHz sampling rate) and randomized into eight complete versions. After being converted back to analog form, the randomized NST lists were re-recorded in a room having variable reverberation time. The reverberation room used for this purpose was a steel-walled enclosure of 6000 cubic feet (20 x 25 x 12 feet) designed to allow for controllable amounts of reverberation by attaching panels of absorptive material to the exposed surfaces of the room. To ensure that the stimuli consisted of primarily reverberant energy, all reverberation measurements and subsequent recordings were completed using a 12-foot loudspeaker-to-microphone distance (the critical distance in this room was between 6 feet and 12 feet). Reverberation was quantified using two different procedures. Manual reverberation measurements were completed by introducing pure tones and narrow bands of noise into the enclosure and measuring the decay pattern displayed on a graphic level recorder. Reverberation time (the time necessary for sound in the room to decay by 60 dB) was extrapolated from the 20-dB or 30-dB decay point. Reverberation was also measured using the Crown Tecron TEF System 10 Audio Spectrum Analyzer. The Energy Time Curve program of this computer-aided system measures reflections in a given frequency band, performs a Fast Fourier Transform on the data, then fits a regression line to the decay pattern to yield a reverberation time value. The stimuli used by the TEF system were sine waves swept over 200-Hz-wide bands centered on octave frequencies between 125 Hz and 8000 Hz. The agreement between reverberation time (RT) values obtained using the two methods decreased as RT increased. Therefore, nominal RT values were calculated by averaging the average RT figures at 500, 1000, and
149--155
March 1990
2000 Hz derived from the two measurement procedures. Nominal RT values used for generating recorded material were 0.6 s for the least reverberant condition, 0.9 s for the moderate reverberation condition, and 1.3 s for the most reverberant condition, representative of RT in small, medium, and large rooms (Lochner & Burger, 1961). One entire NST set was re-recorded twice at each of the three reverberation times: once in quiet, and once in the presence of the cafeteria competition (at +10 dB S:N relative to the word "mark" in the carrier phrase) recorded on the second track of the original NST tapes. Resnick et al. (1975) describe this noise as having a relatively flat power spectrum up to 600 Hz, beyond which energy declines at the rate of 10 dB per octave (see Resnick et al., 1975 for a complete description of this noise). The NST stimuli were played out of a cassette recorder (Sony TC-D5M), amplified (NAD Series 20 Stereo Amplifier), then fed to a loudspeaker (MDM TA-2) placed in the reverberation room at approximately 0° re: a Knowles Electronics Mannequin for Acoustic Research (KEMAR). The exact positioning of KEMAR was determined by measuring the place where the sound pressure level (SPL) outside of KEMAR's two ear canals was the same (this was necessary because of the non-uniform reflection pattern in the reverberation chamber). Two loudspeakers, placed at approximately 450 left and 45 ° right of KEMAR at a distance of 14 feet, transmitted the competition. Stimuli were picked up by microphones (Etymotic Research ER-11) placed in the ear canals of KEMAR, sent through the associated preamplifier designed to remove ear canal resonances (Etymotic Research ER-11), then recorded digitally onto videotape using a Pulse-Code Modulation (Sony PCM 501 ES/NEC VCR VC-N40 EU) recorder. In an attempt to preserve binaural cues the output of KEMAR's right ear was sent to track 1 of the PCM; the left-ear output was routed to track 2. The entire taping procedure was repeated in an anechoic chamber to obtain two lists (one in quiet and one with competition) having essentially no reverberation. Procedures Each subject listened to all eight of the NST lists (0.0, 0.6, 0.9, 1.3 s RT in quiet, and at +10 dB S:N). The NST stimuli were played out of the PCMNCR unit through a clinical audiometer (Grason-Stadler 1701) and were delivered to the listener via TDH-49 headphones. Track 1 of the test tape (with the output from KEMAR's right ear) was sent to the right headphone, and Track 2 of the tape (containing the output from KEMAR's left ear) was routed to the left headphone. The stimuli were delivered at 50 dB re: the pure-tone average, or at the highest level judged comfortable by the participant. Subjects responded by circling their choices on a response form. No feedback was employed, and subjects were instructed to circle a response for each stimulus. Presentation order was randomized across the eight conditions for each subject.
Downloaded From: http://jslhr.pubs.asha.org/ by a Western Michigan University User on 03/27/2016 Terms of Use: http://pubs.asha.org/ss/rights_and_permissions.aspx
HELFER & WILBER: Reverberation and Noise
Subjects Participants were younger (less than 36 years of age) and older (over 60 years of age) ambulatory adults with a negative history of neurological disorder, middle ear pathology, and ototoxic drug usage. In an attempt to obtain a sample of individuals with presbycusis, older subjects were also screened for excessive noise exposure. In addition, none of the subjects demonstrated an asymmetric hearing loss, and all participants learned English as their first language. Participants were divided into four groups, with 8 subjects per group. Table 1 contains age and pure-tone average data for the subject groups. The YN (young normal) group consisted of young adults with no evidence of auditory system disorder. Participants in this group met the following criteria: pure-tone thresholds s20 dB HL (ANSI, 1969) for test frequencies between 250 Hz and 8000 Hz; air-bone gaps .01) for the reverberation plus noise condition for all groups. For reverberation alone, the difference in scores between the two highest reverberation levels was significant only for the OHI group. Thus, although the 1.3 RT recordings were subjectively more distorted than the 0.9 RT lists, the decline in performance from reverberation appeared to plateau at 0.9 RT. A plateau effect in reverberation is also apparent in the data from Nabelek and Robinson (1982) for reverberation alone. Comparison of performance between the OHI and YHI groups is confounded by the difference in amount of hearing loss between these two sets of subjects; discerning aging effects is difficult because of the greater amount of hearing loss in the YHI group. However, other group comparisons were completed using Scheffe post-hoc tests. Performance by the YN group was significantly (p < .01) better than that of the other three subject groups for all NST conditions. Scores obtained by the two groups of older subjects differed significantly (p < .01) for the 0.OQ, 0.ON, 0.9N, and 1.3N conditions; participants in the OM group obtained similar scores to the older adults with greater amounts of hearing loss for stimuli distorted by reverberation alone and mild amounts of reverberation plus noise.
Downloaded From: http://jslhr.pubs.asha.org/ by a Western Michigan University User on 03/27/2016 Terms of Use: http://pubs.asha.org/ss/rights_and_permissions.aspx
153
HELFER & WILBER: Reverberation and Noise
where RScore also showed an inverse relationship with age. Thus, subject age appeared to be an important variable in distorted NST perception, although age was not correlated with hearing loss or with undistorted nonsense syllable perception. Amount of hearing loss was also an important contributor to performance in distortion: both PTA1 and PTA2 were correlated negatively with the average reverberation and reverberation plus noise scores. PTA1 and PTA2 were also correlated highly with each other. NST performance in quiet with no reverberation was correlated positively with MaxRN, RScore, RNScore, and with NST in noise, and correlated negatively with both measures of pure-tone average. When threshold was partialed out the undistorted NST score remained correlated with RNScore, MaxRN, and with NST in noise, although the correlation coefficients were lower. Thus, percent-correct performance on this reverberated nonsense syllable task was related, on an individual basis, to performance on the same task in its undistorted condition. It is not known whether specific consonant error patterns (not analyzed in the present study) were similar for reverberated and unreverberated stimuli. NST distorted by noise alone was correlated negatively with age, PTA1 and PTA2, and correlated positively with undistorted NST, RScore, and RNScore. When PTA1 and PTA2 were partialed out these correlations continued to be significant. The partial correlation matrix also contained a negative correlation between performance in noise alone and percent decline from noise (MaxN).
The percent-decline indices and averaged scores calculated from the NST data were subjected to separate ANOVAs. Highly significant group effects were noted for both RScore [F(3, 28) = 15.745; p < .0001] and for RNScore [F(3, 28) = 21.976; p = .0001]. Post-hoc Scheffe comparisons revealed the YN group to have significantly higher RScore and RNScore values (p < .01), as compared to the other three groups. Group differences in MaxR, MaxN, and MaxRN failed to reach statistical significance; the amount of decline in consonant perception caused by reverberation and noise did not vary systematically among subject groups. CorrelationAnalyses Pearson r and partial correlations were calculated to examine interrelations among the variables. The partial correlations were computed with the two measures of pure-tone average controlled. The correlation matrix is displayed in Table 2. The correlation analyses used the following variables: age, PTA1, PTA2, RScore, RNScore, MaxR, MaxN, MaxRN, NST in quiet with no reverberation (0.0Q), and NST in noise alone (0.ON). The individual NST scores in reverberation and in reverberation plus noise were not utilized for two reasons: first, significant correlations would probably have been obscured because of a relatively large number of variables as compared to the number of subjects; second, the scores on the individual NST lists would have been intercorrelated. Therefore, the average indices (RScore and RNScore) were utilized for all subsequent analyses. The correlation results showed age to be correlated negatively with RNScore and with NST in noise. These correlations persisted in the partial correlation analysis,
DISCUSSION
AND CONCLUSIONS
The majority of older participants in this study, including several individuals with only mild peripheral hearing
TABLE 2. Correlation results for data pooled across all subjects. The upper value in the matrix is the Pearson product-moment coefficient;
the lower value is the correlation coefficient obtained when PTA1 and PTA2 are partialed out (* = statistical significance at the .01 level; **= statistical significance at the .001 level). Age Age
--
NST/Quiet
NST/Quiet -. 292 -. 371 -
-RScore RNScore
RScore
RNScore
NSTINoise
MaxR
MaxN
MaxRN
-. 433* -. 676**
-. 453* -. 631**
.112 .242
.252 .307
.077 .150
-. 025 -
.885**
.869**
.865**
-. 067
.395
.435*
-. 749**
-. 875**
.560** -
.469* .920**
.525** .869**
.392 -. 492*
.423* .159
.469* .135
-. 879**
-. 879**
--
.725**
.609**
-. 460*
-. 099
-. 161
.954**
-. 388
- .026
-.004
.851**
-. 332
-. 445*
-. 438*
-. 304
-. 120
-. 176
-. 549**
-
.423*
.549**
.572**
.689** .690** .675**
-. 255 -. 517*
--
NST/Noise MaxR MaxN
--
--
MaxRN PTA1
PTA2
Downloaded From: http://jslhr.pubs.asha.org/ by a Western Michigan University User on 03/27/2016 Terms of Use: http://pubs.asha.org/ss/rights_and_permissions.aspx
.093 -. 202
PTA1
PTA2 .148
-
-. 758**
-. 861**
-
-. 707** -
.452* -. 188 -.161 -
-. 823** -
.247 -. 223 -. 250 .837**
154 Journal of Speech and HearingResearch
loss, demonstrated marked difficulty understanding nonsense syllables distorted by reverberation and noise. Perhaps the most revealing results was the strong negative correlation between age and performance in reverberation and noise. A connection between age and speech perception in reverberation and noise was also noted by Harris and Reitz (1985) and by Nabelek (1988), especially for conditions with severe amounts of distortion. In fact, in the reverberation plus noise conditions performance by the older hearing-impaired participants was slightly less accurate than that of the younger hearing-impaired listeners, even though pure-tone threshold was correlated with NST scores in distortion and the younger subjects had significantly greater hearing loss. Thus, although peripheral hearing loss was an important contributor to the decline noted in distorted conditions, some factor(s) other than amount of threshold elevation also appeared to be limiting the performance of these older listeners. The reverberation and noise levels employed in this investigation were typical of those encountered in real rooms. This amount of distortion was sufficient to interfere with older adults' ability to perceive consonants. Certainly, the deciphering on nonsense syllables is a difficult, somewhat artificial task, because the listener cannot use linguistic knowledge to fill in distorted or missing portions of the message. At present, it is unknown whether older adults also demonstrate difficulty processing realistic stimuli (i.e., running speech) distorted by reverberation plus noise. The deficit in perception noted in the performance of some of the older adults with minimal hearing loss confirms previous reports of age-related decline in intelligibility of distorted speech. As amount of distortion increases, elderly individuals with little peripheral hearing loss appear to perform less like young, normalhearing listeners (Harris & Reitz, 1985; Nabelek, 1988; Nabelek & Robinson, 1982). Variability, however, is a hallmark in the aging literature, and was observed in our sample of older subjects (several attained scores similar to those obtained by young, normal hearing listeners). Additional research should focus on identifying auditory and non-auditory abilities contributing to age-related speech perception deficits. Correlational analyses revealed a strong connection between pure-tone hearing loss and consonant perception in distortion. A significant relation also existed between both performance in quiet or in noise and NST perception in reverberation plus noise. Because a clinical measure of speech perception in reverberation/noise is not yet available, it would be advantageous to be able to predict performance under realistic listening conditions from data obtained from conventional audiologic tests. Due to the relatively large amount of intersubject variability noted in this and other studies of speech perception in reverberation, it would be unwise to assume that all individuals who perform normally on typical audiologic measures have little difficulty perceiving speech distorted by room acoustics. Although, in this investigation, pure-tone hearing loss and performance on a mea-
33 149-155
March 1990
sure of speech in its undistorted state were related to scores obtained in reverberation/noise, future studies should examine the relation between performance on reverberated and unreverberated versions of other speech material.
ACKNOWLEDGMENTS This article is based upon a doctoral dissertation completed by the first author under the direction of the second author, and was supported in part by a Dissertation Year Fellowship from Northwestern University. We wish to thank Bill Martens, Dean Garstecki, Tom Carrell, and Bill Yost for their assistance on this project and Harry Levitt for supplying the NST tapes. We also thank Rich Freyman, Anna Nabelek, and Peter Fitzgibbons for their comments on an earlier version of the manuscript. Portions of this paper were presented at the American Speech-LanguageHearing Association convention in Boston, Massachusetts, November, 1988. REFERENCES AMERICAN NATIONAL STANDARDS INSTITUTE. (1969). Specificationsfor audiometers (S3.6-1969). New York: ANSI. BERGMAN, M. (1980). Aging and the perception of speech. Baltimore: University Park Press. COOPER, J. C., & CUTTS, B. P. (1970). Speech discrimination in noise. Journal of Speech and HearingResearch, 13, 332-337. CRUM, M. (1974). Effects of reverberation, noise and distance upon speech intelligibility in small, classroom size acoustic enclosures. Unpublished doctoral dissertation, Northwestern University, Evanston, IL. DUBNO, J. R., DIRKS, D. D., & MORGAN, D. E. (1984). Effects of age and mild hearing loss on speech recognition in noise. Journal of the Acoustical Society of America, 76, 87-96. DUQUESNOY, A. J. (1983). The intelligibility of sentences in quiet and in noise in aged listeners. Journalof the Acoustical Society of America, 74, 1136-1144. DUQUESNOY, A. J., & PLOMP, R. (1980). Effect of reverberation and noise on the intelligibility of sentences in cases of presbyacusis. Journal of the Acoustical Society of America, 68, 537-544. GELFAND, S. A., Ross, L., & MILLER, S. (1986). Effects of aging and hearing loss on sentence reception in noise from one or two sources. Journal of the Acoustical of America, 80, s77. HARRIS, R. W., & REITZ, M. L. (1985). Effects of room reverberation and noise on speech discrimination by the elderly. Audiology, 24, 319-324. HAWKINS, D. B., & YACULLO, W. S. (1984). Signal-to-noise ratio advantage of binaural hearing aids and directional microphones under different levels of reverberation. Journal of Speech and Hearing Disorders,49, 278-286. HOUTGAST, T., & STEENEKEN, H. J. M. (1973). The modulation transfer function in room acoustics as a predictor of speech intelligibility. Acustica, 28, 66-73. HUMES, L. E., DIRKS, D. D., BELL, T. S., AHLSTROM, C., & KINCAID, G. E. (1986). Application of the Articulation Index and the Speech Transmission Index to the recognition of speech by normal-hearing and hearing-impaired listeners. Journal of Speech and HearingResearch, 29, 447-462. IRWIN, R. J., & MCCAULEY, S. F. (1987). Relations among temporal acuity, hearing loss, and the perception of speech distorted by noise and reverberation. Journal of the Acoustical Society of America, 81, 1557-1565. ISO (1982). Acoustics-Thresholdof hearingby airconduction as a function of age and sex for otologically normal persons. ISO/DIS 7029. LOCHNER, J. P. A., & BURGER, J. F. (1961). The intelligibility of
Downloaded From: http://jslhr.pubs.asha.org/ by a Western Michigan University User on 03/27/2016 Terms of Use: http://pubs.asha.org/ss/rights_and_permissions.aspx
HELFER & WILBER: Reverberation and Noise speech under reverberant conditions. Acustica, 11, 195-200. MONCUR, J. P., & DIRKS, D. (1967). Binaural and monaural speech intelligibility in reverberation. Journal of Speech and HearingResearch, 10, 186-195. NABELEK, A. K. (1988). Identification of vowels in quiet, noise, and reverberation: Relationships with age and hearing loss. Journalof the Acoustical Society of America, 84, 476-484. NABELEK, A. K., & DAGENAIS, P. A. (1986). Vowel errors in noise
and in reverberation by hearing-impaired listeners. Journal of the Acoustical Society of America, 80, 741-748. NABELEK, A. K., & JENNINGS, E. (1981). Age effects on speech
perception in reverberation. Journal of the Acoustical Society of America, 72, s89. NABELEK, A. K., & LETOWSKI, T. R. (1985). Vowel confusions of hearing-impaired listeners under reverberant and non-reverberant conditions. Journal of Speech and Hearing Disorders, 50, 126-131. NABELEK, A. K., & MASON, D. (1981). Effect of noise and
reverberation on binaural and monaural word identification by subjects with various audiograms. Journal of Speech and Hearing Research, 24, 375-383. NABELEK, A. K., &PICKETT, J. M. (1974). Monaural and binaural
speech perception through hearing aids under noise and reverberation with normal and hearing-impaired listeners. Journalof Speech and HearingResearch, 17, 724-739. NABELEK, A. K., & ROBINETTE, L. (1978a). Reverberation as a
parameter in clinical testing. Audiology, 17, 239-259. NABELEK, A. K., & ROBINETTE, L. (1978b). Influence of the
155
OLSEN, W. O., NOFFSINGER, D., & KURDZIEL, S. (1975). Speech
discrimination in quiet and in white noise by patients with peripheral and central lesions. Acta Otolaryngologica, 80, 375-382. PALVA, A. (1955). Studies of hearing for pure-tones and speech in noise. Acta Otolaryngologica,70, 232-241. PLOMP, R., & DUQUESNOY, A. J. (1980). Room acoustics for the aged. Journal of the Acoustical Society of America, 68, 16161621. PLOMP, R., & MIMPEN, A. M. (1979). Speech-reception threshold for sentences as a function of age and noise level. Journalof the Acoustical Society of America, 66, 1333-1342. RESNICK, J. B., DUBNO, J. R., HOFFNUNG, S., & LEVITT, H.
(1975). Phoneme errors on a nonsense syllable test. Journalof the Acoustical Society of America, 58, (Suppl. 1), 114. Ross, M., HUNTINGTON, D. A., NEWBY, H. A., & DIXOQN, R. F.
(1965). Speech discrimination of hearing-impaired individuals in noise. Journal of Auditory Research, 5, 47-72. SCHUKNECHT, H. (1955). Presbycusis. Laryngoscope, 65, 407-
419. SILMAN, S., & GELFAND, S. A. (1981). The relationship between magnitude of hearing loss and acoustic reflex threshold levels. Journal of Speech and Hearing Disorders, 46, 312-316. SMITH, R. A., &PRATHER, W. F. (1971). Phoneme discrimination
in older persons under varying signal-to-noise conditions. Journal of Speech and Hearing Research, 14, 630-638. WINER, B. J. (1971). Statisticalprinciplesin experimentaldesign (2nd ed.). New York: McGraw-Hill.
precedence effect on word identification by normally hearing and hearing-impaired subjects. Journal of the Acoustical Society of America, 63, 187-194.
Received March 6, 1989 Accepted August 7, 1989
NABELEK, A. K., & ROBINSON, P. K. (1982). Monaural and binau-
ral speech perception in reverberation for listeners of various ages. Journal of the Acoustical Society of America, 71, 12421248.
Request for reprints should be sent to Karen S. Helfer, University of Massachusetts, Department of Communication Disorders, Arnold House, Amherst, MA 01003.
Downloaded From: http://jslhr.pubs.asha.org/ by a Western Michigan University User on 03/27/2016 Terms of Use: http://pubs.asha.org/ss/rights_and_permissions.aspx
