Journal Logo

Research Articles

Understanding Self-reported Hearing Disability in Adults With Normal Hearing

Kamerer, Aryn M.; Harris, Sara E.; Kopun, Judy G.; Neely, Stephen T.; Rasetshwane, Daniel M.

Author Information
doi: 10.1097/AUD.0000000000001161
  • Free
  • Editor's Choice



The prevalence of reported hearing disability (HD) in people with clinically normal hearing is surprisingly high. A survey by Koerner et al. (2020) found that over two-thirds of practicing audiologists encounter at least one patient each month with complaints of a HD despite an audiometric diagnosis of “normal hearing.” More alarmingly, a quarter of audiologists encounter at least four of these patients per month (Koerner et al. 2020). Other surveys of clinical prevalence studies report these patients comprise between 4 and 7% of patients seen in the clinic (Hind et al. 2011; Spehar & Lichtenhan 2018). On a population level, the Beaver Dam Offspring Study reported 12% of people with normal audiometric thresholds complained of hearing difficulties (Tremblay et al. 2015).

As part of a larger ongoing study, we recruited people with and without hearing loss and with and without complaints of HD. We noticed a number of people who, given their audiometric thresholds, were considered to have normal hearing but who participated in the study because they reported a HD. The presented study is a post hoc analysis of self-reported HD and auditory and demographic characteristics of those participants with clinically normal hearing. Proposed explanations for the discrepancy between self-reported HD and audiometric thresholds include, but are not limited to, central presbyacusis, auditory disability with normal hearing, obscure auditory dysfunction, King-Kopetzky Syndrome, auditory dysacusis, auditory processing disorders, idiopathic discriminatory dysfunction, hidden hearing loss or cochlear synaptopathy, auditory neuropathy, inner hair cell damage, strial dysfunction, metabolic disorder, tinnitus, neurocognitive disorders, dyslexia, attention deficit disorder, traumatic brain injury, spatial hearing disorders, aging, dementia, cognitive decline or impairment, receptive aphasia, and Alzheimer’s disease (Beck & Danhauer 2019). The goal of this analysis was not to diagnose any of the aforementioned disorders, but to determine whether a number of demographic variables and common clinical and research measures of auditory function could explain and predict self-reported HD.

A number of surveys and questionnaires have been developed to guide patients in describing their hearing experiences and quantify reported HD. In the present study, we implemented the 12-item Speech, Spatial, and Qualities of Hearing Scale (SSQ12). The SSQ was developed by Gatehouse & Noble (2004) to quantify the degree of disability experienced by a person whose hearing is impaired. The SSQ emphasizes components of auditory scene analysis, which is important to effective communication and quality of life and is directly related to the most common complaint of adults seen in the audiology clinic: difficulty understanding speech in noisy backgrounds. The first section covers several realistic speech hearing contexts of varying difficulty and in different types of background noise; the second section covers three discrete classical components of spatial hearing; direction, distance, and movement; while the third section addresses stream segregation, clarity, and listening effort. The SSQ was validated against audiometric thresholds and a handicap questionnaire with items from the Hearing Disabilities and Handicaps Scale (Hetu et al. 1994) and the Glasgow Benefit Inventory (Robinson et al. 1996). The 12-item version of the SSQ (SSQ12) was developed as a clinical tool that is representative of the original 49-item version (Noble et al. 2013). Compared to several other surveys of HD in adults, the SSQ12 is appropriate for use on people with normal hearing thresholds because it does not ask questions regarding audibility, which is why it has been used in prior research to study self-reported hearing in people with normal audiometric thresholds (Humes et al. 2013).

Predictors of SSQ12 included information obtained from case history (age, sex, history of noise exposure, and tinnitus); three different frequency calculations of PTA and speech intelligibility index (SII) for conversational speech levels; extended high-frequency audiometry (EHF); word recognition scores in quiet, steady-state noise, time compression, and reverberation; frequency-modulation detection thresholds (FMDT); Montreal Cognitive Assessment (MoCA) total score; distortion-product otoacoustic emission (DPOAE) amplitudes, auditory brainstem response (ABR) wave I amplitude; and the frequency following response (FFR) of the speech-evoked ABR.



Data from 111 adults (71 female) with clinically normal hearing between the ages of 19 and 74 (mean = 35, median = 32) years were included in this study. Participant data were collected as part of a larger study that recruited people with and without complaints of difficulty hearing in noisy backgrounds with audiometric thresholds up to 60 dB HL. For data to be included in the present study analyses, participants were required to have thresholds of ≤25 dB HL at all octave and interoctave frequencies between 0.25 and 8 kHz in both ears. In addition, because the SSQ is sensitive to interaural asymmetry (Noble & Gatehouse 2004), to be included in the study participants were required to have symmetric hearing, that is, <10 dB difference between ears at all frequencies between 0.25 and 8 kHz. Furthermore, participants were required to have normal middle ear function, defined as peak pressure within the range of –100 to +50 daPa and static compliance between 0.3 and 2.5 cm3 measured via 226-Hz tympanometry (Otoflex 100, Madsen) and no air-bone gap >10 dB at any octave frequency from 5 to 4 kHz. All additional auditory measures were made monaurally with the test ear selected randomly. In total, there were 43 right ears and 68 left ears included in the study.


Participants completed all measures within two months over two visits. Average data collection time for each participant was approximately four hours. All procedures were approved by the Boys Town National Research Hospital Institutional Review Board, and informed consent was obtained from all participants. Participants were paid for their participation.

12-Item Speech, Spatial, and Qualities of Hearing Scale

All participants completed the 12-item short form of the SSQ (SSQ12; Gatehouse & Noble 2004; Noble et al. 2013). The SSQ12 was designed as a clinical measure of hearing in everyday settings, giving particular attention to speech in competing for sound, localization and directionality of sound, and stream segregation abilities. Each item was given as a situation in which the participant scores their perceived level of ability to perform from 0 to 10. While the response anchors varied by situation, higher scores always reflected greater ability, and lower scores reflected greater disability. The average score across all 12 items was used in analysis.


Pure-tone air conduction thresholds (HDA 200 over-the-ear headphones; Sennheiser, Wedemark, Germany) at octave frequencies (0.25 to 8 kHz) and interoctave frequencies (0.75, 1.5, 3, and 6 kHz) were measured using an audiometer (GSI AudioStar Pro, Grason-Stadler) in 5-dB steps following a modified Hughson-Westlake procedure (ASHA 1978). Individual air conduction thresholds are displayed in Figure 1. Additionally, air conduction thresholds (HDA200 over-the-ear headphones; Sennheiser, Wedemark, Germany) at two extended high frequencies (11.2 and 16 kHz; EHF), were measured using an ascending procedure to avoid discomfort. Equipment limitations only allowed for maximum testing of levels up to 90 dB SPL for 11.2 kHz and 60 dB SPL for 16 kHz. For statistical analysis, participants with no response at maximal levels (“NR” in Fig. 1) were given threshold values of 5 dB above the maximum (e.g., 65 dB SPL for 16 kHz). This was done to minimize missing data (due to a nonresponse) in statistical analyses. In total there were 0 participants with NR to 90 dB SPL at 11.2 kHz and 14 participants with NR to 60 dB SPL at 16 kHz.

Fig. 1.:
Audiometric thresholds for 111 participants. Fourteen participants had no response to a 16 kHz pure-tone at 60 dB HL and are marked no response (NR).

Several derivations of the audiogram were included as predictor variables. Three different frequency-combinations of pure-tone average (PTA) were calculated to assess whether low-, speech range-, or high-frequency thresholds had different relationships with self-reported hearing: a four-frequency average (4FA) over 0.5, 1, 2, and 4 kHz; a low-frequency average (LFA) over 0.5, 1, and 2 kHz, and high-frequency average (HFA) over 4, 6, and 8 kHz. Additionally, a speech-intelligibility index (SII) was calculated to determine whether speech audibility predicted self-reported hearing. The SII for “normal” speech levels of 62.35 dB SPL were calculated according to ANSI S3.5-1997. Input thresholds included all octave and interoctave frequencies from 0.25 to 8 kHz, with interpolation between input frequencies, and critical band SII procedure constants (“sii” package R; Warnes 2013). Only one ear was used for all derivations of the audiogram, as the assumption of symmetry between ears was met for inclusion into the study.

Impulse Noise Exposure

Participants were asked whether they had ever in their lifetime been exposed to impulse noise (e.g., explosion or gunfire) without hearing protection. This single yes/no question has been shown to predict whether a person will have a high or low score on the Lifetime Exposure to Noise and Solvents Questionnaire (Bramhall et al. 2017; Kamerer et al. 2019a,b). Sixty-four participants answered “no” and 47 participants answered “yes.” A large number of participants with high noise exposure is likely due to the targeted recruitment efforts for the larger ongoing study related to noise exposure and is not representative of the population of Omaha, NE.


For the purposes of this study, participants were deemed to have tinnitus if they have experienced ringing, buzzing, or humming for more than two consecutive minutes every day for at least 6 months. Eighty-three participants said “no” and 28 participants said “yes.”

Word Recognition

Stimuli were four 50-word lists spoken by a male talker (NU-6; Auditec, Inc., St. Louis, MO) presented at 65 dB SPL. Word recognition scores (percent correct) for each participant were assessed in four listening conditions: speech in quiet (Quiet), speech in speech-shaped noise presented at 0 dB SNR (Noise), speech in quiet that had been time-compressed by 45% (TC), and speech in quiet that had been time-compressed by 45% and a reverberation time of 0.3 sec (TCR; Noffsinger et al. 1994). Lists and conditions were presented in random order. Performance in each condition was measured as the percent words correct of the final 45 words in each list. The first five words were considered as training to familiarize participants with the condition.

Frequency Modulation Detection

A common behavioral measure of temporal processing ability is modulation-detection (Moore & Glasberg 1989). Here, we assessed temporal processing using a frequency-modulation detection threshold (FMDT; Strelcyk & Dau 2009; Johannesen et al. 2016). This procedure measures the minimum excursion in frequency that a listener can detect. FMDT was measured using a 3-alternative forced-choice (3AFC) adaptive procedure (Yost 1978; Gelfand 1982). For a 3AFC, three intervals were presented with only one interval containing the target stimulus; in this case, one interval contained an FM tone (AudioLab MATLAB; developed by Lopez-Poveda). The participant was required to indicate which interval contained the FM tone and feedback was provided for each response. A 2-up, 1-down adaptive procedure was used to track the 71% point on the psychometric function or threshold.

The stimulus was a pure tone of 1.5 kHz with a duration of 750 ms presented at 70 dB SPL. In one interval, one tone was FM with a variable maximum frequency excursion. The minimum detectable excursion in Hz was estimated and log-transformed. The tones in all three intervals were also sinusoidally-amplitude-modulated with a modulation depth of m = 0.333 or 20log10 [(1+m)⁄(1–m)] = 6 dB (Moore & Glasberg 1989; Johannesen et al. 2016). The low-frequency (1.5 kHz) carrier and amplitude modulation were intended to prevent the participants from using cues based on changes to excitation patterns in the cochlea (Moore & Sek 1996). Following Johannesen et al. (2016), the initial and final modulation frequencies were randomized in the interval between 1 and 3 Hz under the constraint that the FM change was always above 1 Hz. The initial step size of the frequency excursion was log10(1.5). This was decreased to log10(1.26) after four reversals. The adaptive procedure continued until a total of twelve reversals in frequency excursion had occurred. The mean of the last eight reversals was used to determine FMDT (in units of log10 [Hz]). One training run and two additional trials were completed. A trial was excluded and repeated if the SD was >0.2 (Strelcyk & Dau 2009) and/or if the difference between thresholds for the two trials was >0.3. Thresholds from at least two trials, which met the inclusion criteria were averaged.

Montreal Cognitive Assessment

The MoCA was developed as a screening tool for mild cognitive impairment (Nasreddine et al. 2005), and has since established normative data around the world and is commonly used as an estimate of general cognitive function (Rossetti et al. 2011). The MoCA has a maximum score of 30. Normal cognitive function is defined as a score of 26 or higher, mild-cognitive impairment is described by a score of 21 to 25, and scores below 21 are an indication of dementia and Alzheimer’s disease. No participants scored below 21, and thirteen participants scored in the mild cognitive impairment range.

Distortion-Product Otoacoustic Emissions

DPOAEs were measured monaurally using custom software (EMAV, version 3.3; Neely & Liu 1994) following the same protocol as Ridley et al. (2018). In short, two primary tones (f1 and f2) were generated by two separate channels of a 24-bit soundcard (Hammerfall DSP Babyface Pro & Fireface UFX, RME, Germany) routed to sound ports housed in the ER-10X probe microphone system (Etymōtic Research, Elk Grove Village, IL, USA). DPOAE levels were measured at f2 = 1.5 and 4 kHz. The f2/f1 ratio was 1.22. The level of f2 was L2 = 55 dB SPL and the level of f1 was set at L1=61 dB SPL in accordance with Kummer et al. (2000). Stimulus levels were calibrated in-ear and DPOAEs were recorded via the ER-10B+ microphone housed in the probe system. DPOAEs were collected in two separate buffers. The 2f1-f2 frequency bin (resolution of 3.9 Hz) of the two buffers was summed to determine the level of the DPOAE. Noise was calculated by subtracting the contents of the two buffers then averaging the 2f1-f2 frequency bin and the five bins on either side. Data collection ended when one of the three following criteria were met: (1) the noise floor was < –20 dB SPL; (2) artifact-free averaging time was >65.6 sec; or (3) SNR reached 60 dB.

Auditory Brainstem Response

Tone burst ABR

Tone burst-elicited ABR waveforms were recorded at 1.5 and 4 kHz using custom software (Cochlear Response [CResp] version 1.0; Boys Town National Research Hospital, Omaha, NE) on a computer equipped with a 24-bit soundcard (i.e., Babyface Pro or Fireface UFX; RME, Germany). EEG responses were acquired using surface electrodes placed at the forehead (Fpz, ground), vertex (Cz, noninverting active), and an inverting reference electrode placed in the ear canal (ER3-26A gold foil tiptrode). Pure-tones at 1.5 and 4 kHz were gated via Blackman window with duration of 1 ms. Stimuli were presented in alternating polarity monaurally at a rate of 11/sec to an ER-3A insert earphone (Etymotic Research, Elk Grove, IL) connected to the soundcard. The stimulus sound-pressure level (SPL) was 110 dB peak-equivalent SPL. Calibration of the stimulus levels was done using a sound level meter (System 824 and SoundTrack LxT1; Larson Davis, Provo, UT) with the ER-3A connected to the sound level meter via a 2cc coupler (G.R.A.S. 60126, Denmark). High levels were chosen to maximize the number of ABR waves observed in participants, especially those with hearing loss (Ridley et al. 2018), and because larger effects of noise exposure history have been seen at such levels (Bramhall et al. 2017). Electrode impedances were ≤5 kΩ in all cases. The EEG signal was amplified (gain = 100,000), filtered (0.01 to 1.5 kHz; Opti-Amp 8001; Intelligent Hearing Systems, Miami, FL), filtered for line interference using a 60 Hz notch filter, and directed to the computer via the soundcard for averaging. Responses were separated by even and odd recordings and stored in two buffers which were averaged for the final waveform (total averages = 1500 artifact-free responses). Artifact rejection was based on the peak absolute differences between the buffers and was set at a maximum of ±20 μV.

Two examiners independently identified peaks and troughs of ABR wave I. The software allowed for a resolution of 0.02 μV for amplitude and 0.02 ms for latency. The amplitude of wave I was calculated as the difference between the positive peak and the following trough. Latencies were used to clarify disagreements between examiners but were not used for any other analyses. Differences > 0.02 ms, which occurred in 28 of the 224 total waveforms (12.5%), were resolved by a third examiner.

Speech-evoked ABR

The ABR to a speech stimulus was recorded immediately following the tone burst-elicited ABR. A 170-ms synthetic /da/ was chosen as the stimulus because it has been used extensively in complex ABR research (see Skoe & Kraus 2010). The stimulus used in the present study was developed by the Auditory Neuroscience Laboratory at Northwestern University as part of their Brainstem toolbox. The /da/ is a six-formant syllable synthesized at a rate of 20 kHz. The duration is 170 ms, with a voicing onset at 10 ms (100 Hz fundamental frequency). Additional details of the formant frequencies and transitions can be found in Song et al. (2011). The stimulus was played at a level of 90 dB SPL at a rate of 4/sec. The EEG was band-pass filtered at cutoff frequencies of 0.1 to 3 kHz and the processing delay of the soundcard was taken into account when analyzing the data. The analyses performed on the speech ABR were directed at the response of the periodic (vowel) portion of the stimulus, therefore the response to the initial transient portion of the stimulus and formant transition portion were removed and analyses were performed over the portion of the response that was delayed 60 to 170 ms relative to the onset of the stimulus (Song et al. 2011). From the steady-state portion of the response, the FFR was calculated as the strength of the spectral components of the response relative to the noise floor.

Statistical Analyses

First, a probability model was created to explain the variability in the SSQ12 scores of participants with normal audiometric thresholds. Second, the predictors in the final model were tested in their ability to classify participants with and without reported HD.

Explaining variability in the SSQ12

Categorical predictors included sex (M/F), impulse noise exposure (Yes/No), tinnitus status (Yes/No). Continuous predictors included the audiometric variables (4FA, LFA, HFA, and SII), extended high frequency thresholds (11.2k and 16k), word recognition scores (Quiet, Noise, TC, and TCR), MoCA score, DPOAE amplitude (DP1.5k and DP4k), ABR wave I amplitude (I1.5k and I4k), and FFR strength. Model building began with exploratory analyses of the relationship between each predictor and the SSQ12. For categorical variables, a one-way analysis of variance determined differences in SSQ between the binary predictors. For continuous variables, Pearson correlational analyses were performed. Only predictors with significant relationships to the SSQ12 (p < 0.05) were included in the following ordinary least squares regression models. Due to the number of predictor variables tested, p values were adjusted to correct for the family-wise error rate of at least one false positive. We report raw statistics denoted with an asterisk ‘*’ for those that remained significant after the Holm correction.

Numeric variables were centered before inclusion into the models, though the reported model statistics have been reverted back to the raw values to aid in interpretation. The best subset of predictors was found using stepwise regression (olsrr; Hebbali 2018). Ordinary least squares regression is sensitive to missing values, therefore, any participants with missing values were removed from the analyses. The best-fit model was determined by a combination of R2, Adjusted R2, Predicted R2, Akaike Information Criteria (AIC), and Sawa’s Bayesian Information Criteria (SBIC). If the best-fit model included a predictor from a measure that included other related predictors, such as predictors derived from audiometry or DPOAEs at two different frequencies, a set of models substituting all predictors from that measure were compared to determine which predictor from that set resulted in better model fit. If collinearities occurred in predictors across measures, such as auditory thresholds and OAEs, both were kept in the model and an interactive term was included if significant (p < 0.05). The final model was the best-fit model after testing substitutions of predictors derived from the same measure and any potential interactions of collinear variables included.

To accurately break down the variance explained by each predictor variable and account for collinearity between predictors, an assessment of relative importance was implemented on the final model (relaimpo; Grömping 2015). Relative importance can be defined as the proportionate contribution each predictor makes to R2, considering both a direct effect and its effect when combined with other variables in the regression equation (Johnson & LeBreton 2004). This approach is based on sequential sums of squares but accounts for the dependence on ordering, that is biased by correlated predictors, by averaging over orderings.

Classifying people with reported hearing disability

We were interested in whether the final model produced by linear regression could differentiate people with normal hearing thresholds who report HD and those who do not (no-HD). Participants were divided into HD and no-HD groups based on the distribution of SSQ12 scores (Fig. 2). The lowest scorers (red) consisted of 30 participants with scores in the bottom quartile (i.e., ≤25th percentile). SSQ12 scores of the lowest scorers ranged from 2 to 6.3 with a mean score of 5.2. These scores are consistent with the mean score of 59 participants with audiometric thresholds in the moderate hearing loss range (41 to 60 dB HL from 0.25 to 8 kHz) who completed the SSQ12 as part of the larger study from which these data were taken. These scores (mean = 5.5) are also consistent with those reported by older adults with moderate hearing loss (Gatehouse & Noble 2004). On the basis of this, we concluded that the range reported by the lowest scorers was consistent with self-reported HD. The 52 participants with scores between the bottom and top quartiles (i.e., 25th to 75th percentile) ranged from 6.6 to 8.3 with a mean score of 7.5 (Fig. 2, light gray). These scores were consistent with the mean score reported by 25 people with mild hearing loss (26 to 40 dB HL from 0.25 to 8 kHz) from the larger study and reflect a generally positive hearing experience with some difficulty in challenging situations. The highest scorers (Fig. 2, blue) consisted of 29 participants in the top quartile (i.e., ≥75th percentile), with SSQ12 scores ranging from 8.4 to 9.7, with a mean score of 9, indicating a positive hearing experience with little to no trouble hearing in challenging situations (no-HD). The predictors from the final best-fit model were included in a logistic regression model to classify participants with HD. The receiver operating characteristic and area under the curve (AUC) are reported.

Fig. 2.:
Speech, Spatial, and Qualities of Hearing Scale (SSQ12) score for the 111 participants with normal audiometric thresholds compared with participants from the larger ongoing study with mild (n = 25) or moderate (n = 59) hearing loss. The 29 participants who scored in the top quartile of the SSQ (blue) reported little-to-no hearing difficulties and were designated as having no hearing disability (no-HD). The 30 participants with score in the bottom quartile of the SSQ12 (red) had significant self-reported hearing difficulties and were designated as having a hearing disability (HD). The 53 participants scoring in the median quartiles (gray) had some self-reported hearing disability. Boxes represent first, second, and third quartiles, whiskers represent minimum and maximum observations, and dots represent mean values.


This study assessed the ability of a number of clinical measures to explain variability in the SSQ12 scores of 111 people with normal audiometric thresholds. Second, a model of significant predictor measures was tested in its ability to classify participants who had low SSQ12 scores consistent with self-reported HD.

The relationships between the SSQ12 and categorical variables: sex, history of impulse noise exposure, and presence of tinnitus were explored using one-way analysis of variance (Table 1). Sex was not significantly related to SSQ12 (F = 2.70, p = 0.103). Impulse noise exposure was related to SSQ12 (F = 13.93, p < 0.001*), in that participants who reported a history of impulse noise exposure reported lower SSQ12 scores. The relationship between tinnitus and SSQ12 bordered significance (F = 3.69, p = 0.058). The trend in the data showed that participants who reported tinnitus reported lower SSQ12 scores. There were also significant relationships between sex, noise exposure, and tinnitus, as assessed through Chi-squared tests with Yate’s continuity correction (Table 2). There were significantly more males with noise exposure than females (X2 = 17.86, p < 0.001), but no significant differences in tinnitus between males and females (X2 = 1.20, p = 0.273). And there was a significantly higher prevalence of tinnitus in those with a history of noise exposure (X2 = 8.64, p =0.003).

TABLE 1. - ANOVA of SSQ12 and categorical variables
Sum of Squares Mean Square F p
Sex 6.73 6.73 2.70 0.103
Noise Exp. 31.61 31.61 13.93 <0.001*
Tinnitus 9.12 9.12 3.69 0.058
ANOVA, analysis of variance; SSQ12, 12-item version of the Speech, Spatial, and Qualities of Hearing Scale.
*Significant relationship after Holm correction for multiple comparisons. Predictors in bold were included in the subsequent regression models.

TABLE 2. - Number of participants in each demographic
Sex Noise Exp. Tinnitus
Sex M 28 * 12 13 27
F 19 52 15 56
Noise Exp. Y 28 * 19 19 * 28
N 12 52 9 55
Tinnitus Y 13 15 19 * 9
N 27 56 28 55
N, no; Y, yes.
*Significant difference in Chi square.

The relationship between SSQ12 and continuous variables was explored via correlational analyses (Table 3). Measures that were significantly correlated with SSQ12 included all standard audiometric derivations and FMDT. Significant correlations between measures were noted. Expected correlations included metrics derived from the same underlying constructs, such PTA and SII. Other correlations have been established in the literature, such as age effects on ABR wave amplitudes. The measures that significantly correlated with SSQ12 (Table 3, bold) were included in building a model of SSQ12 score.

TABLE 3. - Correlation Matrix of SSQ12 and continuous predictor variables
SSQ Age 4FA LFA HFA 11.2k 16k SII MoCA Quiet Noise TC TCR FMDT DP1.5k DP4k I1.5k I4k FFR
SSQ 1 –0.22 –0.39* –0.37* –0.36* –0.15 –0.15 0.41* 0.10 0.08 0.09 0.06 0.04 –0.40* 0.25 0.19 0.17 0.17 0.06
Age –0.22 1 0.32 0.27 0.44* 0.66* 0.66* –0.40* –0.23 –0.19 –0.19 –0.21 –0.30 0.41* –0.24 –0.22 –0.42* –0.42* –0.27
4FA –0.39* 0.32 1 0.96* 0.66* 0.33 0.33* –0.63* –0.28 –0.27 –0.22 –0.32 –0.27 0.38* –0.37* –0.34* –0.14 –0.13 –0.12
LFA –0.37* 0.27 0.96* 1 0.50* 0.27 0.27* –0.52* –0.26 –0.28 –0.17 –0.33* –0.23 0.33 –0.38* –0.28 –0.12 –0.10 –0.09
HFA –0.36* 0.44* 0.66* 0.50* 1 0.54* 0.54* –0.88* –0.21 –0.05 –0.16 –0.10 –0.28 0.33 –0.22 –0.33* –0.20 –0.25 –0.04
11.2k –0.15 0.66* 0.33 0.27 0.54* 1 0.64* –0.54* –0.16 –0.08 –0.10 –0.14 –0.34* 0.34* –0.20 –0.26 –0.30 –0.38* –0.16
16k –0.28 0.68* 0.49* 0.42* 0.58* 0.64* 1 –0.51* –0.24 –0.25 –0.22 –0.14 –0.22 0.34* –0.19 –0.26 –0.26 –0.28 –0.12
SII 0.41* –0.40* –0.63* –0.52* –0.88* –0.54* –0.54* 1 0.12 0.00 0.16 0.16 0.33* –0.41* 0.25 0.33* 0.18 0.24 0.03
MoCA 0.10 –0.23 –0.28 –0.26 –0.21 –0.16 –0.16 0.12 1 0.33 0.12 0.31 0.24 –0.20 0.10 0.03 0.23 0.15 0.20
Quiet 0.08 –0.19 –0.27 -0.28 -0.05 -0.08 -0.08 0.00 0.33 1 0.35* 0.42* 0.31 -0.11 0.04 0.10 0.03 -0.06 0.16
Noise 0.09 –0.19 –0.22 –0.17 –0.16 –0.10 –0.10 0.16 0.12 0.35* 1 0.17 0.24 –0.24 0.17 0.10 0.01 –0.04 0.18
TC 0.06 –0.21 –0.32* –0.33 –0.10 –0.14 –0.14 0.16 0.31 0.42* 0.17 1 0.45* –0.27 0.01 0.02 0.03 –0.02 0.09
TCR 0.04 –0.30 –0.27 –0.23 –0.28 –0.34* –0.22 0.33* 0.24 0.31 0.24 0.45* 1 –0.33* 0.09 0.11 0.17 0.17 0.19
FMDT –0.40* 0.41* 0.38* 0.33 0.33 0.34* 0.34* –0.41* –0.20 –0.11 –0.24 –0.27 –0.33* 1 –0.11 –0.13 –0.25 –0.21 –0.06
DP1.5k 0.25 –0.24 –0.37* –0.38* –0.22 –0.20 –0.20 0.25 0.10 0.04 0.17 0.01 0.09 –0.11 1 0.47* 0.33 0.34* 0.13
DP4k 0.19 –0.22 –0.34* –0.28 –0.33* –0.26 –0.26 0.33* 0.03 0.10 0.10 0.02 0.11 –0.13 0.47* 1 0.16 0.10 0.14
I1.5k 0.17 –0.42* –0.14 –0.12 –0.20 –0.30 –0.30 0.18 0.23 0.03 0.01 0.03 0.17 –0.25 0.33 0.16 1 0.84* 0.26
I4k 0.17 –0.42* –0.13 –0.10 –0.25 –0.38* –0.28 0.24 0.15 –0.06 –0.04 –0.02 0.17 –0.21 0.34 0.10 0.84* 1 0.13
FFR 0.06 –0.27 –0.12 –0.09 –0.04 –0.16 –0.19 0.03 0.20 0.16 0.18 0.09 0.19 –0.06 0.13 0.14 0.26 0.13 1
FFR, frequency following response; FMDT, frequency-modulation detection thresholds; HFA, high-frequency average; LFA, low-frequency average; MoCA, Montreal Cognitive Assessment; SII, a speech-intelligibility index; SSQ12, 12-item version of the Speech, Spatial, and Qualities of Hearing Scale; 4FA, four-frequency average.
*Significant Pearson correlation after Holm correction for multiple comparisons. Predictors in bold were included in the subsequent regression models.

Accounting for Variance in SSQ12

The initial model included history of impulse noise exposure, PTA-4FA, and FMDT. One participant was missing FMDT and was removed from analysis. Not all audiometric measures were included in the initial model to reduce effects of collinearity on the stepwise procedure. Stepwise regression found a model that included all three predictors yielded the highest predicted R2 and lowest AIC and SBIC values. Next, four models which substituted other metrics derived from the audiogram (LFA, HFA, and SII) were compared. The model that produced the highest Adjusted R2 and F-value included SII. FMDT and SII were correlated in that those with higher SII had lower FMDT (r = –0.41, p < 0.001*); therefore, the inclusion of an interactive term was added. The interaction term FMDT × SII was not significant and thus removed from the final model. The final model, then, included FMDT, impulse noise exposure, and SII, which accounted for 31% of the variance in SSQ (Table 4; F = 16.20, degrees of freedom [df] (3,106), p < 0.0001*).

TABLE 4. - Final model
Including Outliers Excluding Outliers
Est. SE t p Est. SE t p
FMDT –2.02 0.65 –3.09 0.002* –2.16 0.58 –3.76 <0.001*
Noise Exp. (Yes) –0.93 0.26 –3.56 <0.001* –0.94 0.23 –4.07 <0.001*
SII 34.62 11.49 3.01 0.003* 39.22 10.10 3.88 <0.001*
Intercept –24.40 11.57 –2.11 0.037* –28.73 10.18 –2.82 0.006*
R 2 0.31 0.41
Est., estimated coefficients; FMDT, frequency-modulation detection thresholds; SE, standard error; SII, a speech-intelligibility index.
*Significant t-statistic.

Figure 3A plots the predicted and observed SSQ12 scores. The final model predicted the observed SSQ12 scores of the 111 participants with a slope of one; however, there were three outliers in the model prediction that skewed the distribution of the residual error to nonnormality (Fig. 3A, solid boxes). The Shapiro-Wilk statistic was borderline significant (W = 0.977, p = 0.05). Removal of these three outliers resulted in a model with normally distributed residuals that increased the accounted variance in SSQ12 to 41% (Table 4; F = 23.97, df (3,103), p < 0.0001*). Further details on these three outlying participants can be found in the Discussion section. Relative importance analysis revealed that SII accounted for 16%, FMDT 14%, and history of impulse noise exposure 11% of the total 41% explained by the model (Fig. 3B).

Fig. 3.:
(A) Predicted SSQ12 score based on a model including FM detection threshold (FMDT), impulse noise exposure (NoiseExp), and speech intelligibility index (SII), compared with actual SSQ12 score. Dashed line represents unit slope. Solid line plots model equation. Filled boxes represent outliers in the model. (B) Variance in SSQ12 score explained by each predictor (three outliers excluded).

Classifying People With Reported Hearing Disability

Using the lowest and highest quartiles of the SSQ12 score, participants were classified as having self-reported HD or no reported HD (no-HD). Those with HD reported scores consistent with the mean scores of people suffering from moderate hearing loss as determined by audiometric thresholds (Fig. 2 and results reported by Gatehouse & Noble 2004). The final model, that included FMDT, history of impulse noise exposure, and SII was assessed in its ability to classify participants with reported HD. The classification ability of this model was moderate, with a sensitivity of 0.87, and specificity of 0.76, and an AUC of 0.81. When the three outliers of the final model prediction (Fig. 3, filled boxes) were removed, AUC increased to 0.86, sensitivity increased to 0.89 and specificity to 0.86 (Fig. 4).

Fig. 4.:
Receiver operating characteristic (ROC) mean curve (black solid line) and 95% confidence intervals (shaded area), for classifying participants as having hearing disability (HD) or no hearing disability (no-HD).


Data from 111 participants taken from a larger ongoing study of noise exposure and hearing loss were analyzed to explain and predict responses to the SSQ12 using a number of demographics, and behavioral and physiological measures related to auditory function. A model of SSQ12 score was built to explain the variance in SSQ12 score for the sample and predict individual scores. The final model was tested on its ability to distinguish participants with high SSQ12 scores, indicative of minimal complaints of HD, or low SSQ12 scores similar to the HD expressed by people with moderate audiometric hearing loss. A model including history of impulse noise exposure, speech intelligibility index, and FM detection threshold was able to account for approximately 40% of the variance in SSQ12 score and classify participants with self-reported HD with a sensitivity of 89% and specificity of 86%, when three outliers were removed from the data.

Impulse noise exposure and self-reported hearing disability

Noise exposure, especially repeated offenses of high-intensity impulse noise, may damage the cochlea in a way that will not elevate thresholds. The traditional view of noise exposure was that it was thought to primarily damage outer hair cells and their stereocilia (i.e., the cochlear amplifier), resulting in either temporary or, after repeated offenses, permanent shifts in hearing sensitivity depending on level and duration of exposure (Dobie & Humes 2017; Neuberger et al. 1992; Clark & Pickles 1996). In the present study, a history of impulse noise exposure was indeed related to audiometric thresholds (z = 4.45, p = 0.04), with participants who answered ‘yes’ to a history of impulse noise exposure having a PTA-4FA of ~ 5 dB higher than participants who answered ‘no’. Though noise exposure and PTA were related, there was no significant difference in SII between noise exposure groups, and relative importance analysis revealed they added independent information to the model, suggestive of potential additional noise-exposure-related pathology.

A debated present hypothesis suggests that audiometrically undetectable damage from noise exposure, termed “hidden hearing loss,” may manifest as an impaired ability to understand speech in the presence of background noise (Schaette & McAlpine 2011). This hypothesis stems from animal studies which show exposure to noise damages the synapses coupling inner hair cells and auditory nerve fibers (“synaptopathy”), mostly fibers that respond to high sound levels, leaving auditory thresholds intact (Kujawa & Liberman 2009; Furman et al. 2013; Liberman & Liberman 2015). Translation of these findings to the clinic has been fraught with species-specific physiologies, methodological difficulties such as quantifying noise exposure, and lack of control over other influential factors such as genetics and exposure (see reviews by Hickox et al. 2017; Le Prell 2019). A measure that has been found to diagnose synaptopathy in animals is the ABR wave I amplitude. Though not statistically significant, participants with no-HD had larger and less variable ABR wave I amplitudes than the HD group. We cannot conclude that this trend may reveal a direct relationship between auditory nerve health and self-reported hearing difficulties because of the dependence of ABR wave amplitudes on factors such as age, sex, and hearing sensitivity. Females, who dominated the HD scoring SSQ12 group, tend to have larger wave I amplitudes (Dehan & Jerger 1990; Don et al. 1993; Jerger & Hall 1980; Stamper & Johnson 2015a,b,2016). Age and auditory thresholds positively correlate with ABR wave amplitude (Jerger & Hall 1980), even in populations under 40 years of age (Kamerer et al. 2019a,b).

Audiometric thresholds and self-reported hearing disability

We were surprised to find the importance of audiometric measures in explaining variance in the SSQ12 scores of people who have thresholds within normal limits. This inevitably led us to question the origins of the upper limits of what is considered normal, and how they ended up in common practice. The present classification of normal hearing is based on highly debated interpretations of error (e.g., Kryter 1973; Davis 1973; Cohen 1973). Martin and Champlin (2000) provide an excellent summary of how the upper limit of 25 dB HL came about. In summary, the American Standards Association (ASA) originally interpreted the upper limit of normal hearing to be 15 dB HL (American Standards Association 1951). This can be found in the first edition of the “Handbook of Clinical Audiology” by Katz and colleagues (1978). Alongside the ASA limit is a much more generous limit of 26 dB HL presented by the International Standards Organization (Davis & Kranz 1964). The ISO determined their standard based on data collected across countries and using different audiometric equipment. They found a 10 dB difference in normal hearing thresholds across all these studies and interpreted this difference as audiometer calibration error when in actuality the 10 dB difference across studies was due to differences in the scientists’ operating definitions of what normal hearing means. Different operational definitions are also what drive the upper limits proposed by Northern & Downs (2002), Goodman (1965), and Jerger & Jerger (1980) that audiology students will find in later editions of “The Handbook” (Katz et al. 2009). According to the results of the present study, the clinical classification of “normal” hearing as ≤25 or 26 dB HL may be missing a pathologic group of patients. In the present study, we found that participants who reported similar levels of HD as people with moderate hearing loss (HD group) generally had higher thresholds than the no-HD. Figure 5 shows the mean (solid line) and SDs (diagonal lines) of audiometric thresholds for groups with reported HD (red) and no-HD (blue). The SD of the no-HD group tends to stay below 10 dB HL. Ideally, in healthy normal-hearing young adults, thresholds are on average 0 dB HL, with SDs of 3 to 5 dB in the 0.1 to 8 kHz range (Wilber et al. 1988; Han & Poulsen 1998; American National Standards Institute 1996). The 95% confidence intervals dictate the true mean likely lies within 10 dB of the average.

Fig. 5.:
Mean (solid line) and SDs (diagonal lines) of audiometric thresholds for groups with reported HD (red) and no-HD (blue).

Speech Intelligibility Index had a slightly stronger correlation with, and explained more variance in, the SSQ12 than the calculations of PTA; however, calculation of SII is more complex than calculating PTA, especially at the moment during a diagnostic evaluation. If PTA-4FA is substituted in the final model, the variance explained is reduced from 41% to 38%, with the three percent loss due to the difference in relative variance explained by SII and PTA-4FA. This difference is arguably not clinically meaningful, and PTA-4FA could be used to explain SSQ12 scores. In fact, in our sample, there were only two participants in the no-HD group with a PTA-4FA ≥10 dB HL and the PTA limit below which no self-reported HD was found was 0 dB HL.

Clark (1981) argues that classification systems based on the audiogram oversimplify hearing loss and are dangerously overinterpreted by otolaryngologists and audiologists. While classifications may be useful in discussions among clinicians and researchers, they can have detrimental effects on patients’ “acceptance and psychological adjustment to the resulting handicap.” For patients believing they suffer from HD, to be told they have normal hearing can be devastating, because a normal diagnosis implies no suggestion for treatment or remediation. The misinterpretation of data surrounding normal dB HL values, and zeal for the classification of hearing loss by audiometric thresholds, has resulted in adopting a classification of “normal hearing” that does not align with subjective reports of HD. Humes (2021) argues for the use of classifications of “auditory wellness,” that relies not on audiometric thresholds but rather self-report to capture both sensory impairment and effects of other hearing disabilities that may not be measured by the audiogram. The results of the present study suggest the SSQ12 might also be of use as a measure of auditory wellness, capturing both audiometric and nonaudiometric factors contributing to HD.

Temporal processing and self-reported hearing disability

The degradation of the ability to encode temporal fine structure in speech can exacerbate difficulty understanding speech in noisy situations (Bharadwaj et al. 2015). An FM detection task was specifically included in the larger study, from which these data were analyzed, to measure temporal fine structure coding. The ability of participants to recognize small changes in frequency over time was different between HD and no-HD SSQ12 groups, with the HD group having significantly higher detection thresholds than participants who reported little-to-no difficulty hearing. While in theory, this may suggest a pathology like noise-induced synaptopathy in the no-HD group, FMDT was not related to impulse noise exposure history in these participants. We have found in a prior study that the behavioral nature of the FM detection task itself may produce such differences between participants (Kamerer et al. 2019a,b). Working memory capacity is highly associated with this specific FMDT task. During the task, the participant is listening for one of three intervals that contain an FM sound. Participants are instructed on what differentiates an FM tone, but as modulation depth decreases it is increasingly difficult to hear clear FM. It is possible that eventually, the task devolves into an odd-one-out task, where participants are listening for the interval that is different from the other two. This type of task requires the storage of the information in all three intervals and processing of this information before the recall. This, in fact, is the definition of a working memory task. Another explanation is that effective temporal processing does require working memory, processing speed, and attention –a relationship that has been shown by some measures of working memory such as operation and symmetry span (Broadway and Engle, 2011) but not reading span (Fullgrabe et al. 2015). Because we did not include a direct measure of working memory in the present study, the relationship between FMDT and self-reported HD could actually be an indirect indicator of differences in working memory between these participants. While we did include MoCA scores in the analysis, the working memory-related questions of the MoCA (1) constitute only a portion of the survey and (2) are intended as a screener for cognitive impairment, not sensitive to small differences in the abilities of cognitively healthy people. Further studies should include FMDT or other temporal processing measures, and a sensitive measure of working memory to separate the contributions to self-reported hearing difficulties.

The three outliers

Diagnostics of the model residuals found three outliers whose SSQ12 scores were poorly predicted by FMDT, impulse noise exposure, and SII. Two of these outliers reported more HD than predicted and one reported much less disability than predicted. It is important to take a closer look at the characteristics of these participants to understand why the model over- or under-predicted their disability. In the two participants who reported HD, the primary finding was that they scored more zeroes on SSQ12 items than any other participants. One outlying participant marked low scores (two zeroes) for the three questions related to sound localization, while marking higher scores for the questions related to speech and qualities of hearing. The other participant marked low scores for almost every question, which could be an error, for example, if the participant flipped the extremes. We did not attempt to contact the participants for further investigation. The third outlier scored much higher on the SSQ12 than predicted by the model. This participant was also one of the older participants sampled (74 years old), and while they had low audiometric thresholds (and therefore high SII) and did not have a history of impulse noise exposure, they had a high FMDT, which caused the poor model fit.

Other examined variables and self-reported hearing disability

We were surprised to find that age was not a significant predictor of SSQ12 score, nor was it significantly different between the HD and no-HD groups (z = -1.65, p = 0.099), though there was a trend toward younger adults reporting higher scores. A number of studies have shown age effects potentially related to how one’s perspective on the degree of reported HD changes over the lifetime compared with changes in auditory thresholds, but the results of such studies are not always in agreement. Some studies have shown that younger adults underestimate their hearing abilities while older adults are better estimators of their audiometric thresholds, but can also overestimate their abilities (Kamil et al. 2015; Oosterloo et al. 2020; Hämäläinen et al. 2021), yet others have shown that older adults with normal hearing thresholds report more disability hearing in difficult situations than younger adults (Banh et al. 2012; Gatehouse & Noble 2004).

The word recognition tasks utilized in this study did not capture self-reported HD either. In spite of presenting words in difficult listening conditions that drove performance below ceiling, we were unable to find differences between the highest and lowest scorers. Our results, like other studies, demonstrated that people with thresholds below 25 dB HL perform highly on word recognition tasks, maintaining the notion that 25 dB HL is within normal hearing limits. However, we demonstrated that people with thresholds ≥10 dB HL reported disability similar to people with moderate hearing loss. The lowest scoring question on the SSQ12 in the present study was Q2 “You are listening to someone talking to you, while at the same time trying to follow the news on TV. Can you follow what both people are saying?”, relating to speech streaming; however, our words-in-noise task did not correlate with total SSQ12 score, nor did it correlate with the speech-related questions of the SSQ12 (r = 0.06, df = 109, p = 0.487), a result corroborated by Lopez-Poveda et al. (2017). This may be evidence that either word-level clinical tests are not capturing real-world difficulties, or that the SSQ12 is overestimating a speech-in-noise deficit. This study did not implement any speech-in-speech or sentence recognition tasks, which may be more sensitive to subclinical pathologies. Monson and colleagues (2019) found that when speech maskers are recorded from a microphone at a 90° azimuth, speech-in-speech recordings capture a more real-world listening scenario than when maskers are recorded from a microphone at 0° azimuth, and that EHF hearing may assist in stream segregation in this acoustical environment (Hunter et al. 2020; Monson et al. 2019). Considering the significant variability in EHF thresholds in this study sample, we might see differences in performance between HD and no-HD groups on speech-in-speech recognition tasks if they are recorded in this manner. While we did not find a significant correlation between EHF thresholds and self-reported disability, Zadeh et al. (2019) did find that people who reported difficulty hearing in background noise had higher EHF thresholds than those who did not report difficulty.

Limitations of the Study

Quantifying self-reported HD is a difficult task. Though 41% of the variance in SSQ12 was explained by the demographic and auditory variables included in the present study, a large amount of variance goes unexplained. Because the data for the analysis were taken from a sample of a larger prospective study on the effects of noise exposure on measures of hearing, measures used in this analysis were not prospectively chosen. This is why DPOAEs were collected only at two stimulus frequencies and more common electrophysiological metrics such as click-evoked ABR and ABR threshold were not included, and why measures of central auditory processing and cognition, and consideration of personality traits were not included in the study.

The SSQ12 item which had the largest difference in score between HD and no-HD was Question 12 “Do you have to concentrate very much when listening to someone or something,” related to listening effort. The mean SSQ12 score for the HD group was 3.6 compared to 9.1 for the no-HD group. These results confirm the conclusions of another recent study surveying self-reported hearing difficulties in adults with normal hearing, who found that complaints involving the presence of background noise were three times higher than complaints involving the audibility and spatial qualities of the signal (Pang et al., 2019). Similar to the findings of this study, Pang and colleagues also found 60% of participants selected “I need to pay a lot of attention and concentrate on the speaker to follow conversations,” as the first or second description of their hearing difficulties. Listening effort has been studied fervently in psychoacoustics research for the past decade, but less focused on the relationship between listening effort and self-reported HD in people with clinically normal hearing. Bologna et al. (2013) found that reported listening effort across young normal-hearing subjects during a speech-in-speech task was independent of their performance on the task. Future studies should consider measures of listening effort which may provide insight into self-reported HD in otherwise healthy adults. In addition to listening effort and the cognitive constructs underlying it, some of the remaining variances could be explained by personality traits such as anxiety that have been linked to reported hearing difficulties, typically referred to as King-Kopetzky syndrome or Obscure Auditory Dysfunction (Gatehouse, 1991; Hinchcliffe, 1992; Saunders & Haggard, 1992; Zhao & Stephens, 1996).

Another limitation of the study was that complex relationships between measures of hearing were, for the most part, ignored. For example, because we did not recruit based on participant demographics, there was a sex imbalance across the noise exposure groups (Table 2). Although sex was included in the analysis, sex is so highly correlated with noise exposure in this cohort (most low-noise participants were female and most high-noise participants are male), adjusting for sex may have skewed the noise exposure effect. To analyze all moderating and/or mediating relationships between these variables, we would need a much larger sample size.


  1. History of impulse noise exposure, audiometric thresholds, and FM detection threshold predicted scores on the SSQ12 in people with audiometric thresholds ≤25 dB HL.
  2. A logistic-regression classifier including history of impulse noise exposure, speech intelligibility index, and FM detection threshold achieved moderate performance (AUC = 0.86) in predicting whether people scored in the top 25% of SSQ12 scored (i.e., reported no-HD) or the bottom 25% (i.e., reported HD similar to people with moderate hearing loss).


We thank Mark Chertoff for his consultation in writing and review of the manuscript.


American National Standards Institute. (1996). American National Standard Specification for Audiometers. (ANSI S3.6-1996).
American Standards Association. (1951). American Standard Specification for Audiometers for General Diagnostic Purposes. (A24.5-1051).
ASHA. (1978). Guidelines for manual pure-tone threshold audiometry. ASHA, 20, 297–301.
Banh J., Singh G., Pichora-Fuller M. K. (2012). Age affects responses on the Speech, Spatial, and Qualities of Hearing Scale (SSQ) by adults with minimal audiometric loss. J Am Acad Audiol, 23, 81–91.
Beck D. L., Danhauer J. L. (2019). Amplification for adults with hearing difficulty, speech in noise problems - and normal thresholds. J Otolaryngol-ENT Res, 11, 84–88.
Bharadwaj H. M., Masud S., Mehraei G., Verhulst S., Shinn-Cunningham B. G. (2015). Individual differences reveal correlates of hidden hearing deficits. J Neurosci, 35, 2161–2172.
Bologna W. J., Chatterjee M., Dubno J. R. (2013). Perceived listening effort for a tonal task with contralateral competing signals. J Acoust Soc Am, 134, EL352–EL358.
Bramhall N. F., Konrad-Martin D., McMillan G. P., Grist S. E. (2017). Auditory brainstem response altered in humans with noise exposure despite normal outer hair cell function. Ear Hear, 38, e1–e12.
Broadway J. M., Engle R. W. (2011). Individual differences in working memory capacity and temporal discrimination. PLoS One, 6, e25422.
Clark J. G. (1981). Uses and abuses of hearing loss classification. ASHA, 23, 493–500.
Clark J. A., Pickles J. O. (1996). The effects of moderate and low levels of acoustic overstimulation on stereocilia and their tip links in the guinea pig. Hearing Res, 99, 119–128.
Cohen A. (1973). Some general reactions to Kryter’s paper “Impairment to hearing from exposure to noise”. J Acoust Soc Am, 53, 1235–1236.
Davis H., Kranz F. W. (1964). The international standard reference zero for pure-tone audiometers and its relation to the evaluation of impairment of hearing. J Speech Hear Res, 7, 7–16.
Davis H. (1973). Some comments on “Impairment to hearing from exposure to noise” by K. D. Kryter. J Acoust Soc Am, 53, 1237–1239.
Dehan C. P., Jerger J. (1990). Analysis of gender differences in the auditory brainstem response. Laryngoscope, 100, 18–24.
Dobie R. A., Humes L. E. (2017). Commentary on the regulatory implications of noise-induced cochlear neuropathy. Int J Audiol, 56(Suppl 1), 74–78.
Don M., Ponton C. W., Eggermont J. J., Masuda A. (1993). Gender differences in cochlear response time: An explanation for gender amplitude differences in the unmasked auditory brain-stem response. J Acoust Soc Am, 94, 2135–2148.
Fullgrabe C., Moore B. C. J., Stone M. A. (2015). Age-group differences in speech identification despite matched audiometrically normal hearing: Contributions from auditory temporal processing and cognition. Front Aging Neurosci, 6, 1–25.
Furman A. C., Kujawa S. G., Liberman M. C. (2013). Noise-induced cochlear neuropathy is selective for fibers with low spontaneous rates. J Neurophysiol, 110, 577–586.
Gatehouse S. (1991). The contribution of central auditory factors to auditory disability. Acta Oto-Laryngologica, 111(Suppl 476), 182–188.
Gatehouse S., Noble W. (2004). The speech, spatial and qualities of hearing scale (SSQ). Int J Audiol, 43, 85–99.
Gelfand S. A. (1982). Hearing: An introduction to psychological and physiological acoustics. J Neurol Neurosurg Psychiatry, 45. doi: 10.1136/jnnp.45.12.1175-b
Goodman A. (1965). Reference zero levels for pure tone audiometer. Am Speech Hear Assoc, 7, 262–263.
Grömping U. (2015). Relative importance for linear regression in R : The Package relaimpo. J Stat Softw, 17. doi: 10.18637/jss.v017.i01
Han L. A., Poulsen T. (1998). Equivalent threshold sound pressure levels for Sennheiser HDA 200 earphone and Etymotic Research ER-2 insert earphone in the frequency range 125 Hz to 16 kHz. Scand Audiol, 27, 105–112.
Hebbali A. (2018). olsrr.
Hetu R., Getty L., Noble W., Stephens D. (1994). Mise au point d ’ un outil clinique pour la mesure d ’ incapacites auditives et de handicaps (Development of a Clinical Tool for the Measurement of the Severity of Hearing Disabilities and Handicaps). J Speech-Language Pathol Audiol, 18, 83–95.
Hickox A. E., Larsen E., Heinz M. G., Shinobu L., Whitton J. P. (2017). Translational issues in cochlear synaptopathy. Hear Res, 349, 164–171.
Hämäläinen A., Pichora-Fuller M. K., Wittich W., Phillips N. A., Mick P. (2021). Self-report measures of hearing and vision in older adults participating in the canadian longitudinal study of aging are explained by behavioral sensory measures, demographic, and social factors. Ear Hear, 42, 814–831.
Hinchcliffe R. (1992). King-Kopetzky syndrome: An auditory stress disorder. J Audiol Med, 1, 89–98.
Hind S. E., Haines-Bazrafshan R., Benton C. L., Brassington W., Towle B., Moore D. R. (2011). Prevalence of clinical referrals having hearing thresholds within normal limits. Int J Audiol, 50, 708–716.
Humes L. E. (2021). An approach to self-assessed auditory wellness in older adults. Ear Hear, 42, 745–761.
Humes L. E., Kidd G. R., Lentz J. J. (2013). Auditory and cognitive factors underlying individual differences in aided speech-understanding among older adults. Front Syst Neurosci, 7, 55.
Hunter L. L., Monson B. B., Moore D. R., Dhar S., Wright B. A., Munro K. J., Zadeh L. M., Blankenship C. M., Stiepan S. M., Siegel J. H. (2020). Extended high frequency hearing and speech perception implications in adults and children. Hear Res, 397, 107922.
Jerger J., Hall J. (1980a). Effects of age and sex on auditory brainstem response. Arch Otolaryngol, 106, 387–391.
Jerger J., Jerger S. (1980b). Measurement of hearing in adults. M. Paparella, Shumrick D. (Eds.), In: Otolaryngology (2nd ed.). W.B. Saunders.
Johannesen P. T., Pérez-González P., Kalluri S., Blanco J. L., Lopez-Poveda E. A. (2016). The influence of cochlear mechanical dysfunction, temporal processing deficits, and age on the intelligibility of audible speech in noise for hearing-impaired listeners. Trends Hear, 20, 2331216516641055.
Johnson J. W., LeBreton J. M. (2004). History and use of relative importance indices in organizational research. Organizational Res Meth, 7, 238–257.
Kamerer A. M., Kopun J. G., Fultz S. E., Allen C., Neely S. T., Rasetshwane D. M. (2019). Examining physiological and perceptual consequences of noise exposure. J Acoust Soc Am, 146, 3947.
Kamerer A. M., AuBuchon A., Fultz S. E., Kopun J. G., Neely S. T., Rasetshwane D. M. (2019). The role of cognition in common measures of peripheral synaptopathy and hidden hearing loss. Am J Audiol, 28, 843–856.
Kamil R. J., Genther D. J., Lin F. R. (2015). Factors associated with the accuracy of subjective assessments of hearing impairment. Ear Hear, 36, 164–167.
Katz J., Chasin M., English K. M., Hood L. J., Tillery K. L. (Eds.). (1978). Handbook of Clinical Audiology (1st ed.). Williams & Wilkins.
Katz J., Medwetsky L., Burkard R., Hood L. J. (Eds.). (2009). Handbook of Clincal Audiology (6th ed.). Lippincott Williams and Wilkins.
Koerner T. K., A Papesh M., Gallun F. J. (2020). A questionnaire survey of current rehabilitation practices for adults with normal hearing sensitivity who experience auditory difficulties. Am J Audiol, 29, 738–761.
Kryter K. D. (1973). Reply to the critiques of A. Cohen, H. Davis, B.L. Lempert, and W.D. Ward of the paper “Impairment to hearing from exposure to noise”. J Acoust Soc Am, 53, 1244–1252.
Kujawa S. G., Liberman M. C. (2009). Adding insult to injury: Cochlear nerve degeneration after “temporary” noise-induced hearing loss. J Neurosci, 29, 14077–14085.
Kummer P., Janssen T., Hulin P., Arnold W. (2000). Optimal L1−L2 primary tone level separation remains independent of test frequency in humans. Hearing Res, 146, 47–56.
Le Prell C. G. (2019). Effects of noise exposure on auditory brainstem response and speech-in-noise tasks: A review of the literature. Int J Audiol, 58(Suppl 1), S3–S32.
Liberman L. D., Liberman M. C. (2015). Dynamics of cochlear synaptopathy after acoustic overexposure. J Assoc Res Otolaryngol, 16, 205–219.
Lopez-Poveda E. A., Johannesen P. T., Pérez-González P., Blanco J. L., Kalluri S., Edwards B. (2017). Predictors of hearing-aid outcomes. Trends Hear, 21, 2331216517730526.
Martin F. N., Champlin C. A. (2000). Reconsidering the limits of normal hearing. J Am Acad Audiol, 11, 64–66.
Monson B. B., Rock J., Schulz A., Hoffman E., Buss E. (2019). Ecological cocktail party listening reveals the utility of extended high-frequency hearing. Hear Res, 381, 107773.
Moore B. C. J., Glasberg B. R. (1989). Mechanisms underlying the frequency discrimination of pulsed tones and the detection of frequency modulation. Effects of carrier frequency, modulation rate, and modulation waveform on the detection of modulation and the discrimination of modulation type. (a. Citation: J Acoust Soc Ame, 86, 2468. doi: 10.1121/1.411967
Moore B. C., Sek A. (1996). Detection of frequency modulation at low modulation rates: Evidence for a mechanism based on phase locking. J Acoust Soc Am, 100(4 Pt 1), 2320–2331.
Nasreddine Z. S., Phillips N. A., Bédirian V., Charbonneau S., Whitehead V., Collin I., Cummings J. L., Chertkow H. (2005). The Montreal Cognitive Assessment, MoCA: A brief screening tool for mild cognitive impairment. J Am Geriatr Soc, 53, 695–699.
Neely S. T., Liu Z. (1994). BOYS TOWN NATIONAL RESEARCH HOSPITAL EMAV: Otoacoustic Emission Averager.
Neuberger M., Körpert K., Raber A., Schwetz F., Bauer P. (1992). Hearing loss from industrial noise, head injury and ear disease. A multivariate analysis on audiometric examinations of 110,647 workers. Audiology, 31, 45–57.
Noble W., Gatehouse S. (2004). Interaural asymmetry of hearing loss, Speech, Spatial and Qualities of Hearing Scale (SSQ) disabilities, and handicap. Int J Audiol, 43, 100–114.
Noble W., Jensen N. S., Naylor G., Bhullar N., Akeroyd M. A. (2013). A short form of the Speech, Spatial and Qualities of Hearing scale suitable for clinical use: The SSQ12. Int J Audiol, 52, 409–412.
Noffsinger D., Wilson R. H., Musiek F. E. (1994). Department of Veterans Affairs compact disc recording for auditory perceptual assessment: Background and introduction. J Am Acad Audiol, 5, 231–235.
Northern J. L., Downs M. P. (2002). Hearing in Children (5th ed.). Lippincott Williams and Wilkins.
Oosterloo B. C., Homans N. C., Baatenburg de Jong R. J., Ikram M. A., Nagtegaal A. P., Goedegebure A. (2020). Assessing hearing loss in older adults with a single question and person characteristics; Comparison with pure tone audiometry in the Rotterdam Study. PLoS One, 15, e0228349.
Pang J., Beach E. F., Gilliver M., Yeend I., (2019). Adults who report difficulty hearing speech in noise: An exploration of experiences, impacts and coping strategies. Int J Audiol, 58, 851–860.
Ridley C. L., Kopun J. G., Neely S. T., Gorga M. P., Rasetshwane D. M. (2018). Using thresholds in noise to identify hidden hearing loss in humans. Ear Hear, 39, 829–844.
Robinson K., Gatehouse S., Browning G. G. (1996). Measuring patient benefit from otorhinolaryngological surgery and therapy. Ann Otol Rhinol Laryngol, 105, 415–422.
Rossetti H. C., Lacritz L. H., Cullum C. M., Weiner M. F. (2011). Normative data for the Montreal Cognitive Assessment (MoCA) in a population-based sample. Neurology, 77, 1272–1275.
Saunders G. H., Haggard M. P. (1992). The clinical assessment of “Obscure Auditory Dysfunction” (OAD) 2. Case control analysis of determining factors. Ear Hear, 13, 241–254.
Schaette R., McAlpine D. (2011). Tinnitus with a normal audiogram: Physiological evidence for hidden hearing loss and computational model. J Neurosci, 31, 13452–13457.
Skoe E., Kraus N. (2010). Auditory brain stem response to complex sounds: A tutorial. Ear Hear, 31, 302–324.
Song J. H., Nicol T., Kraus N. (2011). Test-retest reliability of the speech-evoked auditory brainstem response. Clin Neurophysiol, 122, 346–355.
Spehar B., Lichtenhan J. T. (2018). Patients with normal hearing thresholds but difficulty hearing in noisy environments : A Study on the Willingness to Try Auditory Training. Otol Neurotol, 39, 950–956.
Stamper G. C., Johnson T. A. (2015a). Auditory function in normal-hearing, noise-exposed human ears. Ear Hear, 36, 172–184.
Stamper G. C., Johnson T. A. (2015b). Letter to the editor: Examination of potential sex influences in. Auditory function in normal-hearing, noise-exposed human ears, ear hear, 36, 172-184. Ear Hear, 36, 738–740.
Stamper G. C., Johnson T. A. (2015). Letter to the editor: Examination of potential sex influences in Stamper G.C, Johnson T.A. (2015). Auditory function in normal-hearing, noise-exposed human ears, Ear Hear, 36, 172-184 Greta. Ear Hear, 36, 738–740.
Strelcyk O., Dau T. (2009). Relations between frequency selectivity, temporal fine-structure processing, and speech reception in impaired hearing. J Acoust Soc Am, 125, 3328–3345.
Tremblay K. L., Pinto A., Fischer M. E., Klein B. E., Klein R., Levy S., Tweed T. S., Cruickshanks K. J. (2015). Self-reported hearing difficulties among adults with normal audiograms: The beaver dam offspring study. Ear Hear, 36, e290–e299.
Warnes G. (2013). Calculating Speech Intelligibility Index (SII) using R, (March). (pp. 1–33).
Wilber L. A., Kruger B., Killion M. C. (1988). Reference thresholds for the ER-3 A insert earphone. J Acoust Soc Am.
Yost W. A. (1978). A forced-choice adaptive procedure for measuring auditory thresholds in children. Behavior Res Meth Instrum, 10, 671–677.
Zadeh L., Silbert N. H., Sternasty K., Swanepoel W., Hunter L. L., Moore D. R. (2019). Extended high-frequency hearing enhances speech perception in noise. Proc Natl Acad Sci USA, 116, 23753–23759.
Zhao F., Stephens D. (1996). Hearing complaints of patients with King-Kopetzky syndrome (obscure auditory dysfunction). Br J Audiol, 30, 397–402.

Audiometry; Hidden hearing loss; Listening difficulties; Noise; Self-report

Copyright © 2021 Wolters Kluwer Health, Inc. All rights reserved