Perceptions of Vocal Performance Impairment in Singers with and without Hearing Loss

Abstract

Introduction.

It is well understood that hearing plays an important role in accurate vocal production. Singers in particular rely on auditory cues and auditory feedback to reproduce specific pitch contours. Therefore, even mild hearing loss may have a detrimental effect on a singer’s ability to perform. This study investigates the effect of hearing loss on self-reported vocal production in singers, specifically in the domains of pitch matching, pitch maintenance, vowel production, and dynamic control as well as the effect of hearing loss on vocal handicap.

Methods.

An 18-item electronic self-report survey was distributed to the members of the National Association of Teachers of Singing and to the Voice Foundation. Data collected included demographics, the Singing Voice Handicap Index-10 (SVHI-10), and a series of closed and open-ended questions. Demographic variables, variables related to the SVHI-10, and variables related to the newly introduced survey were included in a hierarchical regression analysis to determine significant relationships.

Results.

Among 206 eligible participants, 37 individuals reported a voice problem, 58 reported hearing loss, and 19 reported concurrent hearing loss and a voice problem. Among males, there were no significant differences between hearing impaired and normal hearing singers in reported pitch matching, pitch maintenance, dynamic control, and vowel matching when those with voice problems were excluded and included. However, in females, when singers with voice problems were excluded, there was a significant difference between hearing impaired and normal hearing singers in pitch matching (P = 0.38). Additionally, when singers with voice problems were included in the female subset, significant differences emerged between the hearing impaired and normal hearing singers in areas of pitch matching (P = 0.01) and vowel matching (P = 0.02). Further, controlling for gender, when excluding voice problems, there was a significant difference between the SVHI-10 scores of normal hearing (mean = 9.03) and hearing impaired participants (mean = 11.30, P = 0.02). This difference continued to be significant when including those with voice problems (normal hearing mean = 9.97, hearing impaired mean=14.1, P <0.0001). Additionally, individuals with hearing impairments were more likely to report higher perceived vocal handicap scores as refiected on the SVHI-10 than normal hearing respondents (P = 0.002). Other factors associated with higher likelihood of SVHI-10 score include older age (P = 0.008), having a voice problem (P <0.0001), and being paid to sing within the past six months (P = 0.001).

Conclusion.

When controlling for voice problems, singers with hearing impairments subjectively did not perceive that they performed less accurately on pitch matching, pitch maintenance, dynamic control, and vowel matching, yet they scored higher on the SVHI-10 indicating vocal handicap. Further study is needed to characterize the relationship between perceived and measured vocal accuracy in singers with hearing loss.

INTRODUCTION

It is well understood that good hearing plays an important role in vocal production.¹ Professional and avocational singers are particularly dependent on their hearing as they need to be able to hear background music, match frequencies, and follow fellow singers when performing.^2,3 These needs generally exceed those required for normal speech and conversation.³ In singers, while slight variations in vocal production may not affect their communication, these affect their ability to sing accurately.² Even small changes in pitch, vowel intonation, and volume or intensity matching can have a detrimental effect on a singer’s ability to adequately perform.²

Some effects of hearing loss on speech and voice production in the general population have been studied. Patients with congenital deafness have difficulty acquiring and maintaining normal speech, and may have difficulty with voicing, intonation, voice quality, and vowel substitutions.^3,4 Due to this, speech in these individuals is often difficult to understand compared to the speech of normal hearing individuals.^4,5 Additionally, individuals who experience hearing loss later in life often have deterioration in their speech over time with greater variability in pitch and difficulty with articulation.^4,6

However, there is little known about hearing loss among singers specifically. Studies in singers with normal hearing have shown that acute changes in auditory input affect vocal output in terms of reactionary pitch and phoneme errors. Pitch matching accuracy decreases amongst singers when external acoustic feedback is masked,^4,7–9 and the magnitude of this effect is greater when the difficulty of the vocal task increases.⁷ However, the effect of prolonged hearing loss in singers has not been identified. Further, these simulated conditions cannot accurately capture the more nuanced difficulties that may be experienced by a singer with intrinsic hearing loss in a more naturalistic musical environment. Additionally, pitch accuracy is not the only component of vocal production that can potentially be affected by hearing loss. Existing studies among singers do not address factors such as vocal timbre, vowel production, and dynamic control, all of which are important for high-quality singing performance.

This survey study was conducted to investigate the effect of hearing loss on subjective vocal difficulties experienced by singers, specifically in the domains of pitch matching, pitch maintenance, vowel production, and dynamic control. Further, we assessed whether singers with hearing loss have greater singing voice handicap and vocal dissatisfaction than the general singing population as measured by the SVHI-10.

METHODS

Population

The study participants consisted of adults who self-identified as singers (professional or avocational). Recruitment of participants took place via members of the National Association of Teachers of Singing and The Voice Foundation. Participants were invited to complete an anonymous survey to examine the role hearing plays in vocal production accuracy among singers. Individuals who were younger than 18 years old or did not self-identify as singers were excluded. The study was reviewed as exempt by the University of Southern California Institutional Review Board.

Survey

This self-report survey collected information on the participants’ singing background such as voice type, time spent singing weekly, and level of formal training. The survey also measured the respondents’ perception of vocal production currently and at their best using a visual analog scale, specifically in the domains of pitch matching, pitch maintenance, dynamic control, and vowel matching. Participants provided additional information regarding their degree of hearing loss, hearing amplification device used, and their voice problems. Respondents who indicated that they have hearing loss were also asked a binary question regarding whether hearing loss affected their vocal production in the domains identified above. Demographic data collected included age, gender, and race. No identifiable data was collected. Additionally, the Singing Voice Handicap Index-10 (SVHI-10) was used to assess the participants’ perceived handicap associated with singing voice problems. The SVHI-10 is an abbreviated version of the 36 item Singing Voice Handicap Index that measures the impact of dysphonia among singers. Both indices have been validated in a diverse population of singers; the SHVI-10 has the additional advantages of high test-retest reliability and reduced question number, decreasing the burden on participants of completing a longer measure. Responses to the SVHI-10 are recorded using a five point Likert scale (0-never, 1-almost never, 2-sometimes, 3-almost always, 4-always).^10,11

The survey consisted of an online questionnaire developed on Qualtrics. The principal investigator drafted the first iteration of the survey. The development process then advanced to further refinement with input from laryngologists and speech pathologists of the USC Voice Center, a senior auditory scientist, and a Thornton School of Music professor of vocology from the University of Southern California. Consensus was achieved amongst the members of the review team. The survey was then pilot-tested amongst six participants matching the target population and further refined before implementation. Participants completed the 18-item self-report survey; participants who reported current hearing loss were asked to complete an additional 7 items.

Data analysis

Data was analyzed using Stata statistical software (StataCorp, College Station, Texas, USA). Data were analyzed for statistically significant relationships among variables using chi-squared test for discrete variables and Student t-test for continuous variables. Mann Whitney test was used for nonparametric data. The level of significance was set at p < 0.05. Due to the large predominance of females in the participant population, the perceived effect of hearing loss on vocal production was analyzed separately for males and for females. Participants’ perceived vocal handicap was analyzed using SVHI-10 scores. A higher score indicates more singing handicap. In order to facilitate inclusion in a logistic regression model, SVHI scores were dichotomized such that participants scoring >=10 were considered to have perceived vocal handicap. The cutoff value of 10 was chosen because it is just above the confidence interval of normative SVHI-10 scores in professional singers (7.43–9.34),^10–12 and it is the median of the score distribution in this study. Variables with univariate p-values <0.10 were considered for inclusion in a logistic regression model with SVHI-10 score >= 10 as the dependent variable. A backwards stepwise logistic regression was conducted to arrive at the final odds ratio model.

RESULTS

Two hundred and twenty four participants anonymously completed the survey from October 20, 2020 through November 30, 2020. Five participants were excluded because they did not self-identify as singers and 13 were excluded for not completing the survey. Table 1 shows the demographic data for the participants. The average age is 44.9. One hundred and fifty nine participants identify as females, 43 identify as males, and 4 identify as other. Seven voice types are represented with Soprano being the most common (38.8%) (Table 2).

TABLE 1.

Demographic Data of the Study Population (n = 206)

Age (mean ± standard deviation) in years	44.9 ± 16.5
Range (minimum to maximum)	21 to 86 years
Gender
Male	43 (20.8%)
Female	159 (77.2%)
Other	4 (1.9%)
Hearing Problem
Yes	58 (28.2%)
Mild	28 (48.3%)
Moderate	22 (37.9)
Profound	2 (3.4%)
Severe	5 (8.6%)
No	148 (71.8%)
Voice Problem
Yes	37 (18.0%)
No	169 (82.0%)

TABLE 2.

Singing background data of the study population (n = 206)

Hours Spent Singing per Week
0–1	12 (5.8%)
2–3	36 (17.5%)
4–5	43 (20.9%)
6–7	29 (14.1%)
8–9	14 (6.8%)
10–11	18 (8.7%)
12–13	11 (5.3%)
14–15	16 (7.8%)
16–17	9 (4.4%)
18–19	4 (1.9%)
20+	13 (6.3%)
No response	1 (0.49%)
Formal Vocal Training?
Yes	187 (90.8%)
No	19 (9.2%)
Professional Singer?
Yes	134 (65.1%)
No	72 (34.9%)
Voice type
Alto	14 (6.8%)
Baritone	14 (6.8%)
Bass	4 (1.9%)
Contralto	3 (1.5%)
Mezzo-Soprano	59 (28.6%)
Soprano	80 (38.8%)
Tenor	25 (12.1%)
Other	6 (2.9%)
Not Sure	1 (0.49%)
Perfect Pitch?
Yes	28 (13.6%)
No	178 (86.4%)

Of the 206 eligible participants, 37 individuals reported they have a voice problem. Fifty-eight individuals reported having hearing loss, the majority of which was reported as mild (48.3%) or moderate (37.9%) hearing loss (Table 1). Nineteen respondents reported concurrent hearing loss and a voice problem. Those with a hearing problem were found to be significantly more likely to have voice problems than normal hearing respondents (P = 0.001) (Figure 1). Further, participants who reported a hearing problem were more likely to be older (median age 56.5 years) than those without a hearing problem (median age 38 years) (P <0.0001). When asked to rate their accuracy on a visual analog scale, participants perceived themselves as very accurate in terms of pitch matching, pitch maintenance, dynamic control, and vowel matching (Figure 2a). Among males, there was no significant difference between hearing impaired and normal hearing singers when singers with voice problems were included and when they were excluded. In fact, the majority of male singers with hearing loss reported that they do not think their hearing loss negatively impacts their voice in pitch matching (82.3%), pitch maintenance (88.2%), dynamic control (82.3%), and vowel matching (94.1%). Among females, when singers with voice problems were excluded, there was a significant difference between the hearing impaired and normal hearing respondents in pitch matching (normal hearing=88.5%, hearing impaired = 96%, P = 0.038). Further, when those with voice problems were included, significant differences emerged between the hearing impaired and normal hearing singers in areas of pitch matching (normal hearing = 87%, hearing impaired = 95.5%, P = 0.01) and vowel matching (normal hearing = 88%, hearing impaired = 98%, P = 0.02) (Figure 2b). Overall, females with hearing loss tended to rate their accuracy as lower than their normal-hearing counterparts while males with hearing loss tended to rate their accuracy as higher than their normal hearing counterparts.

FIGURE 1.

Proportion of respondents with voice problems based on hearing loss status.

FIGURE 2.

A. Perceived vocal difficulty in males and females across voice domains in singers with hearing loss and singers with normal hearing, excluding singers with voice problems. B. Perceived vocal difficulty in males and females across voice domains in singers with hearing loss and singers with normal hearing, including singers with voice problems. Box plots represent the interquartile range (IQR) of the 25th and 75th percentile scores with the median value in the center. The horizontal lines extending from the IQR represent the minimum and maximum values and the points beyond the line represent the extreme outliers within the data. Significance was determined using the median values with the Mann Whitney U test. Asterisk indicates significance.

Controlling for gender, when those with voice problems were included, there was a significant difference between the SVHI-10 scores of normal hearing (mean = 9.97) and hearing impaired participants (mean = 14.1, P <0.0001). Further, when those with voice problems were excluded, there continued to be a significant difference between the SVHI-10 scores of normal hearing (mean = 9.03) and hearing impaired participants (mean=11.30, P = 0.02). On univariate analysis, having a voice problem (P <0.0001), having a hearing problem (P = 10 (Figure 3). Age, formal voice training, voice type, and having perfect pitch were not found to be significant. In a multivariate logistic regression model, individuals with hearing impairments were more likely to report SVHI-10 score >= 10 than normal hearing respondents (P = 0.002, OR=4.01, [CI 1.68–9.54]). Other factors independently associated with higher likelihood of SVHI-10 score >= 10 include older age (P = 0.008, OR = 0.97, [CI 0.95–0.98]), having a voice problem (P <0.0001, OR = 43.11, [CI 5.42–342.93]), and being paid to sing within the past six months (P = 0.001, OR = 0.31, [CI 0.16–0.61]).

FIGURE 3.

Proportion of respondents with SVHI-10 score >= 10 and <10 in normal hearing and hearing impaired respondents.

DISCUSSION

This survey study investigated the effect of hearing loss on self-identified singers’ perception of their vocal production in terms of pitch matching, pitch maintenance, dynamic control, and vowel matching. It also assessed singers’ perception of their voice through the use of a self-report survey and the SVHI-10.

Among females, there was a significant difference in the perceived accuracy of pitch matching between the hearing impaired and normal hearing groups when singers with voice problems were excluded. However, among males, there were no significant differences in the perceived accuracy of pitch matching, pitch maintenance, dynamic control, and vowel matching between the hearing impaired and normal hearing groups when including and when excluding those with voice problems. Further, the majority of male hearing impaired singers reported that they do not think their hearing loss negatively impacts their accuracy in any of these domains. This is surprising in light of previous studies that show that using white noise to mask auditory input in singers is associated with a measurable decrease in accuracy across all of these domains.^4,7–9 This discrepancy in our findings may be explained by the idea that there is a hearing loss threshold at which singing difficulties begin to manifest and our respondents may be below that threshold. In fact, of the participants who reported having hearing loss, the majority of the participants reported mild (48.3%) or moderate (37.9%) hearing loss, indicating that they may very well be below the threshold at which hearing loss has a measurable effect on vocal performance. Further prospective studies are warranted to identify exactly what this threshold could be. Additionally, the significant difference in perceived pitch accuracy between female hearing impaired and normal hearing groups in the context of the large predominance of females in our population may indicate that the male subset is simply too underpowered to detect vocal impairment caused by hearing loss.

Interestingly, among females, when singers with existing voice problems were included in the analysis, statistically significant differences emerged between hearing impaired and normal hearing singers in the areas of pitch matching and vowel matching. Though the difference is somewhat small, even small deficits can pose serious problems in singers as they are required to be very accurate and have little room for vocal error. It is expected that vocal pathology would interfere with pitch matching accuracy, since the larynx is directly involved in controlling pitch. However, the effects of vocal problems on vowel production accuracy are less intuitive to explain, since laryngeal pathology would not be expected to directly affect vocal tract shaping. While it is not possible to know from this study, a potential explanation could be that increased vowel matching difficulty in singers with hearing loss may not entirely be explained by laryngeal function impairment.

Further, singers with hearing impairment were more likely to report vocal handicap than normal hearing respondents when controlling for other factors, as measured by higher scores on the SVHI-10. This suggests an important association between these two pathologies. One possible explanation is that hearing loss can itself lead to vocal problems over time by mechanisms yet to be determined. This association has been demonstrated in the literature. In a cross-sectional study of age and gender matched older adults, it was found that individuals with hearing impairment have significantly greater vocal dissatisfaction and were more likely to be diagnosed with a voice disorder.¹³ Similarly, a study of patients with cochlear implants concluded that this population has less effective auditory feedback and poorer vocal precision than normal hearing individuals.⁴

The juxtaposition of increased SVHI-10 scores with high vocal accuracy self-ratings in participants with hearing loss suggests a possible discrepancy between perceived and actual vocal handicaps present in singers with hearing loss. One possibility is that singers with hearing loss are simply less able to hear their own vocal deficits, particularly if these deficits developed slowly over a long period of time. Alternatively, these singers may perceive their vocal deficits, but attribute it to causes other than hearing loss. This explanation is less likely, since hearing-impaired singers did not show significant differences from normal-hearing singers when reporting their perceived vocal accuracy on a visual analog scale. It is also possible that the voice handicap captured on the SVHI-10 refiect impairments in factors that were not queried otherwise in the survey, and does not represent a true difference in perceived and actual vocal handicap.

These potential findings reveal an important therapeutic indication. If hearing impairment truly does contribute to the development of vocal impairment, select singers with vocal difficulties may benefit from aural rehabilitation, such as hearing aid or insert earphone use. This could potentially help singers retain vocal skills in areas such as pitch matching, pitch maintenance, dynamic control, and vowel matching by improving auditory feedback. Further, using therapeutic interventions to diminish the negative effect of hearing loss on vocal production may not only improve a singer’s ability to perform, but also their quality of life. Losing the ability to perform may decrease singers’ sense of self as singing may be an integral part of their identity.

This study has notable limitations. Data was collected using an online survey for a brief period of time. This did not allow for random sampling. A longitudinal collection of data would have allowed for a more accurate analysis and understanding of causal relationships. Also, the self-report nature of the survey measures may have introduced response bias. Further, since the survey only recorded data at one point in time, there was a risk of recall bias that may have infiuenced participant responses. Most importantly, because of the survey-based nature of this study, objective measurements of hearing loss or vocal production accuracy were not included; therefore we were not able to compare the participants’ perceptions to objective measures of hearing loss and vocal production accuracy. In order to confirm whether hearing loss is truly associated with altered vocal production, audiometric data will have to be collected alongside acoustic measures of the singing voice. In addition to assessing fundamental frequencies of sung pitches, attention should be paid to acoustic information in the first and second formant ranges, important for vowel formation, as well as frequencies in the range of the so-called “Singer’s Formant” (3–4 KHz), thought to be especially important in the voice quality of male operatic singers.

Future prospective studies are warranted to identify the specific mechanism by which hearing loss leads to vocal problems in order to develop therapeutic interventions both for prevention and treatment. Further, it is important to characterize the relationship between perceived and actual voice impairments in singers with hearing loss and determine the extent to which performance is affected. Additionally, further research is needed to identify the threshold at which hearing loss becomes severe enough to potentially affect vocal performance.

Biographies

Gurmehr Brar is a 2nd year medical student at Loyola University Chicago, Stritch School of Medicine. She received a Bachelor of Science degree in Global Health from the University of Southern California and a Master of Science degree in Global Medicine from the University of Southern California Keck School of Medicine. She is interested in Otolaryngology and plans to pursue a residency in the field.

Einav Silverstein is a 4th year medical student and current Dean’s Research Scholar at the University of Southern California Keck School of Medicine. Before enrolling at Keck she attended Oberlin College and Conservatory, where she obtained a Bachelor of Arts in Biology and Neuroscience as well as a Bachelor of Music in Vocal Performance. As a Dean’s Research Scholar, her research focus is on the relationship between hearing loss and vocal health in singers. She plans to pursue a residency in otolaryngology, and is particularly interested in the fields of laryngology and voice disorders.

Melissa Zheng is a resident physician in the Department of Otolaryngology-Head and Neck Surgery at the University of Southern California. She completed undergraduate education at Swarthmore College and medical school at the University of California, San Francisco.

M. Eugenia Castro is a speech-language pathologist at the USC Voice Center. She received her master’s degree in Communication Disorders and Sciences and her Bachelor of Arts in music from California State University, Northridge and completed a fellowship in voice, upper airway, and swallowing disorders USC Voice Center. Additionally, she has a degree in vocal performance from Escuela de Música Contemporánea - Berklee International Network in Argentina and has worked as a professional singer. Her clinical and research interests include voice disorders in professional and non-professional voice users, and she specializes in the care of the professional and performing voice.

Dr. Ray Goldsworthy received his Bachelor of Science in Physics in 1997 from the University of Kentucky and his Doctor of Philosophy in Health Sciences and Technology in 2005 from Harvard University and the Massachusetts Institute of Technology. Ray joined the University of Southern California as an Associate Professor of Otolaryngology in 2014 where his research program is dedicated to music appreciation and hearing loss. Ray is a cochlear implant user and is passionate about the interplay of auditory experience, auditory perception, and medical bionics for improving the lives of people with hearing loss.

Lynn Helding, Professor of Practice in Voice and Vocal Pedagogy at USC’s Thornton School of Music, is the author of The Musician’s Mind: Teaching, Learning & Performance in the Age of Brain Science, the chapter “Brain” in Scott McCoy’s Your Voice: An Inside Viewand an associate editor of the Journal of Singing. Helding was recognized as a “legendary figure in the field of voice pedagogy” by the Contemporary Commercial Music Vocal Pedagogy Institute at Shenandoah University, receiving their 2020 Lifetime Achievement Award.

Dr. Michael M. Johns is a graduate of Johns Hopkins School of Medicine. He completed his residency in Otolaryngology at the University of Michigan. He then pursued a fellowship in laryngology and care of the professional voice at the Vanderbilt Voice Center at Vanderbilt University. He is the director of the USC Voice Center at the University of Southern California, pursing research, teaching and clinical care, with a specific interest in laryngeal dystonia, geriatric laryngology and the aging voice.