In audiology, assessing a patient’s hearing ability typically involves measuring their hearing thresholds (HTs), which indicate the softest sounds they can detect. However, it is also important to determine the intensity at which sounds become uncomfortably loud. Known as loudness discomfort levels (LDLs) or uncomfortable loudness levels (UCLs/ULLs), these measurements define the upper limits of the comfortable listening range, also referred to as the dynamic range.

LDLs are typically measured using an ascending method on an audiometer with pure tones of different frequencies, where the intensity is increased in 5 dB steps until discomfort is reported (Punch et al., 2004). Alternatively, LDLs can be measured using noise or speech stimuli, either live or recorded, and less commonly with natural recorded sounds. LDLs are considered to be as reliable as pure-tone HTs, with test-retest differences of <5 dB (Punch et al., 2004; Sherlock and Formby, 2005; Vidal et al., 2022). The British Society of Audiology has published recommended procedures and important caveats regarding LDL measurement to improve consistency and reliability (British Society of Audiology, 2022). Notably, there is no universally standardized method for measuring LDLs and the use of various psychophysical approaches, stimulus types and instructions can lead to different results (Punch et al., 2004). In normal-hearing adults without a sound tolerance problem, pure tone LDL values typically range from 75 to 120 dB HL, with an average of 100 dB HL and a standard deviation between 10.67 and 13.58 dB, showing minimal variation across frequencies (Sherlock and Formby, 2005).

LDLs are useful for setting the maximum power output of hearing aids and for assessing loudness disorders such as hyperacusis (Punch et al., 2004; Sheldrake et al., 2015). Hyperacusis is a condition in which sounds cause physical discomfort or pain at levels that are generally tolerable for most people (Adams et al., 2021; Henry et al., 2022). There is no gold standard measure to clinically assess hyperacusis (Fackrell et al., 2017). While individuals with hyperacusis often have lower LDLs (Aazh and Moore, 2017a; Anari et al., 1999; Enzler et al., 2021; Sheldrake et al., 2015), no universal cut-off criteria have been established. Some recommend reporting low LDLs below 90 dB HL at two frequencies between 0.25 and 8 kHz (Goldstein and Shulman, 1996), while others suggest an average of 77 dB HL or less at 0.25–8 kHz for the worse ear (Aazh and Moore, 2017a). Sheldrake et al. (2015) reported an average LDL in 381 hyperacusis patients of 85 dB HL across 0.125–8 kHz, with values ranging from 30 to over 120 dB HL, highlighting that LDLs lack sensitivity and specificity for diagnosis. Thus, low LDLs do not always indicate hyperacusis, nor do “normal range” LDLs exclude it. LDLs also show small correlations with other hyperacusis assessment methods, such as questionnaire scores (Aazh and Moore, 2017a; Jüris et al., 2013; Meeus et al., 2010; Zaugg et al., 2016).

LDLs are typically measured at some or all standard audiometric frequencies (0.25–8 kHz), which may limit their diagnostic scope. To date, no studies have reported LDLs for extended high frequencies (EHFs; >8 kHz) in either normal or pathological populations. Yet, EHF HTs are crucial for detecting early cochlear damage caused by aging, ototoxicity, or noise exposure (Lough and Plack, 2022) and may also help identify early auditory deficits associated with tinnitus, often described as subclinical damage or hidden hearing loss (Jafari et al., 2022). This means that in tinnitus, HTs may appear normal at standard frequencies but show impairment at EHFs (Basile et al., 2013). Lower LDLs at standard frequencies have also been shown to distinguish adolescents with tinnitus from those without, suggesting LDLs may serve as a sensitive indicator of early auditory dysfunction (Sanchez et al., 2016). Since both tinnitus and hyperacusis are thought to result from increased auditory central gain following peripheral damage (Knipper et al., 2013), hyperacusis may similarly involve distinct patterns at EHFs, either in HTs or LDLs.

While previous studies have found small or non-significant associations between hyperacusis and audiogram configuration at standard frequencies (Brandy and Lynn, 1995; Hannula et al., 2011), or between LDLs and HTs in hyperacusis (Aazh and Moore, 2017a; Anari et al., 1999; Sheldrake et al., 2015), the role of EHFs has not yet been investigated.

This study aims to fill this gap by (1) establishing normative LDL values for extended high frequencies (EHFs; >8 kHz) in a healthy, non-hyperacusis population, (2) examining whether LDLs vary with gender or tinnitus status and their correlations with age and questionnaire scores for hyperacusis, noise sensitivity, anxiety, and depression, and (3) exploratorily presenting LDL data for a small subset of participants who self-report hyperacusis. Older age, female sex, tinnitus, noise sensitivity, anxiety, and depression have all been previously reported as factors associated with hyperacusis (Bigras, Theodoroff, et al., 2024; Smit et al., 2021). Given the lack of normative LDL values for EHFs, this study will also provide insights into assessing comfort levels, familiarization tones, and masking effects in EHF audiometry (Lough and Plack, 2022).