Association Between Age-Related Sensory and Metabolic Hearing Loss and Cognitive Performance

Abstract

The underlying mechanisms linking age-related hearing loss (HL), and cognitive dysfunction are not well-understood. Traditionally, age-related HL was primarily related to damage of hair cells (i.e., the sensory component of HL) however, growing evidence suggests that the endocochlear potential (i.e., the metabolic component of HL) may also play a role. In this study, we investigated the relationship between the sensory and metabolic components of HL and cognitive dysfunction in 100 older adults. Forty participants were patients with mild cognitive impairment recruited during visits to the Memory Clinic at the Danish Dementia Research Centre, while 60 participants were recruited from the general public. All participants were assessed with pure-tone audiometry and a comprehensive cognitive test battery including but not limited to tests examining, episodic memory, processing speed, and executive functions. The sensory and metabolic components of the HL were estimated based on the age-related HL components model. Metabolic HL was strongly correlated with performance on tests of processing speed, moderately correlated with performances on tests of executive function, but not significantly correlated with episodic memory performance. On the other hand, sensory HL was not significantly correlated with any of the cognitive tests. When correcting for age, the correlation between metabolic HL and processing speed remained significant. Thus, participants with lower processing speed were significantly more likely to have a larger metabolic HL. These data could lend support to the idea that shared vascular pathophysiological mechanisms are a key link between age-related HL and cognitive decline.

Keywords

cognition metabolic hearing loss sensory hearing loss hearing loss aging

Introduction

The number of people living with significant cognitive dysfunction was estimated to be 57 million globally in 2019, and at the time it was further estimated that the number will increase to 153 million in 2050, mainly due to increased life expectancy in developing countries (Livingston et al., 2024; Nichols et al., 2022). This does not only result in a large societal cost that in 2019 was estimated to be US$ 1.3 trillion (Wimo et al., 2023), but it also profoundly impacts the quality of life of the families of the patients (Shah et al., 2024). Several cognitive functions including processing speed, attention, executive functions and memory decline with age (Brito et al., 2023; Harada et al., 2013). The degree of decline in cognitive function varies in severity from normal cognition to mild cognitive impairment (MCI) to dementia where MCI is considered a transitional stage between normal aging and dementia (Petersen, 2016; Winblad et al., 2004), with preserved everyday life and functional activities, but with increased risk of developing dementia (Petersen, 2016). It is therefore important to gain a better understanding of how various risk factors affect cognition and how cognitive decline can be prevented.

A growing amount of literature suggests an association between age-related hearing loss (HL) and poorer cognitive performance in older adults (Cantuaria et al., 2024; Deal et al., 2017; Gallacher et al., 2012; Gurgel et al., 2014; Lin et al., 2011; Loughrey et al., 2018; Powell et al., 2022; Strutt et al., 2022). The reported effect size has been small, but nevertheless consistent, across several studies taking place in different parts of the world and using different methodologies (Alattar et al., 2020; Golub et al., 2017; Gurgel et al., 2026; Sugawara et al., 2011). This association has resulted in HL being named as one of the largest potential modifiable risk factors for dementia in midlife (Livingston et al., 2024). A recent population-based cohort study reported that individuals with HL who were not using their hearing aids (HAs) had a higher risk of dementia than those who had HL and were using HAs (Cantuaria et al., 2024). However, a recent randomized controlled trial investigating the use of HAs as a preventive measure against cognitive decline suggests that there is, in fact, no impact of hearing intervention on preventing cognitive decline (Lin et al., 2023). Nevertheless, a secondary analysis in a sub-sample of the participants revealed a potential preventative effect of the hearing intervention in those that were already at a higher risk of developing dementia. Whether and to what extent a causal relationship between the HL and cognitive decline exists is however still not understood. Both indirect mechanisms such as social isolation (Livingston et al., 2020) and cognitive load (Sweller, 2011), as well as shared pathophysiological mechanisms such as inflammation, vascular dysfunction or mitochondrial impairments (Kostrikov et al., 2025) have been suggested. It is therefore necessary to better characterize the association between HL and cognitive dysfunction to be able to better understand the potential role of hearing interventions in the future.

Traditionally, age-related HL was subcategorized into four major categories based on audiometric tests and temporal bone pathology: sensory, neural, metabolic and cochlear conductive presbycusis (Schuknecht & Gacek, 1993). In recent years, the primary focus of research has been on sensory and neural HL and understanding how damaged sensory hair cells or synaptopathy affect hearing in aging. Less focus has been on metabolic HL, which is HL related to a dysfunctional and degraded stria vascularis (SV; Schuknecht & Gacek, 1993). SV is an epithelial structure that lines the lateral wall of the cochlea, and it plays an important role in maintaining the endocochlear potential by transporting potassium ions (K⁺) into the endolymph (Bovee et al., 2024). The endocochlear potential is necessary for the mechanotranduction taking place in the sensory hair cells (Chan & Hudspeth, 2005) and is therefore crucial for perceiving sound. It is not possible to examine the SV non-invasively in humans. However, previous studies based on human histopathological and animal studies have suggested that the audiogram shape can provide information about the etiology of age-related HL (Dubno et al., 2013; Vaden et al., 2016, 2018). Moreover, the extent of the sensory and metabolic components of HL can be estimated for an individual by finding the linear combination of the canonical audiogram shapes that best approximates the individual's audiogram (Vaden et al., 2022). Although audiograms cannot fully capture the pathological issues of the cochlea and auditory pathway, the estimates of the sensory and metabolic components give an indication of the underlying pathophysiology of the HL.

The aim of the present study was first to investigate possible differences in the sensory and metabolic HL components between the MCI patients and the healthy older adults, and second to characterize the association between the sensory and metabolic components of HL and cognitive performance. This investigation was part of a larger study exploring HL and MCI, and therefore, includes older adults both with and without MCI. Pure-tone audiometry and a comprehensive cognitive test battery examining episodic memory, processing speed, and executive functions, as well as other cognitive domains, were measured. The sensory and metabolic components of the HL were estimated based on the age-related HL components model proposed by Vaden et al. (2022).

Methods

Participants

This study was part of a larger study investigating the association between hearing and cognition among older individuals with and without MCI. Thus, results from the same study population have previously been reported in (Hendel et al., 2026). One hundred older adults were recruited between January and July 2024 either from the Copenhagen Hearing and Balance Center, Rigshospitalet, Denmark or the Memory Clinic at the Danish Dementia Research Centre, Rigshospitalet, Denmark. All participants gave informed consent before participating in the study. The study was approved by the Ethics Committee of the Capital Region of Denmark (F-23054924).

Forty participants with MCI were recruited during visits to the Memory Clinic at the Danish Dementia Research Centre, Rigshospitalet and were either newly referred patients or patients visiting for a follow-up appointment. All the MCI patients underwent thorough assessment including physical and neurological examination, blood tests, neuroimaging and cognitive testing. Participants were required to be 55 years of age or older, native Danish speakers and have a clinical syndrome diagnosis of MCI. The MCI diagnosis was defined based on recommendations from Winblad et al. (2004) with the following criteria: (1) not normal function, but not diagnosed with dementia; (2) evidence of cognitive decline as measured by self-reports and deficits on objective cognitive tests, operationalized as performance below −1.5 SD from age and education-adjusted normative data; and (3) activities of daily function mainly preserved. Furthermore, patients recruited in the MCI group were required to have obtained a Mini-Mental State Examination (MMSE) score greater than or equal to 24 (Folstein et al., 1975) and a score corresponding to a Clinical Dementia Rating (CDR) of 0.5 based on an interview about their daily functioning (Hughes et al., 1982). Exclusion criteria included psychiatric disorders, neurogenetic diseases, normal pressure hydrocephalus or a history of either ear surgery or treatment of head and neck cancer.

The second group of participants included 60 cognitively intact older adults recruited through an advertisement in a local newspaper or through the research participant database at the Danish Dementia Research Centre. Inclusion criteria included an age of 55 years or older and being a native Danish speaker. Exclusion criteria included psychiatric or neurological diseases, history of ear surgery or treatment of head and neck cancer, or use of medication that can affect cognition. Furthermore, participants were excluded if their daily functioning interview score corresponded to a CDR score above 0 and/or their MMSE score was below 26.

Questionnaire

A case history questionnaire was used to collect information about demographic characteristics including sex, age and self-report HA use.

Cognitive Tests

All cognitive tests were administered by a trained cognitive rater or a neuropsychologist and took place either in a sound attenuated room to control for noise disturbances or in the Memory Clinic. All verbal instructions and stimuli were at suprathreshold levels and hearing aid users wore their HAs during the examination.

Fluency

Both a lexical and a category fluency test were included to examine language, tempo and executive functions (Lezak et al., 2004). The lexical test consisted of three rounds where the participants were asked to name as many words as possible in 1 minute starting with either F, A, or S. The total FAS score is the summed amount of all words named except proper nouns from all three rounds. In the category test, participants were asked to name as many animals as possible within one minute and the score is the number of correct words. All animals were accepted, regardless of category and subcategory.

Logical Memory (LM)

In the test, a short story was presented orally, and participants were asked to verbally recall as many details as possible about the story immediately after the presentation (immediate recall) and again approximately 25 minutes after the first recall (delayed recall). Only part A of the test was administered resulting in a score from 0 to 25 points for each recall condition based on the number of details remembered. The test evaluated verbal episodic memory and was based on a subtest of the Wechsler Memory Scale (Wechsler, 1997).

Reys Complex Figure Test (RCFT)

The test consisted of two parts. In the first part, the participants were asked to copy a complex line drawing in freehand when the figure was presented for the first time (RCFT copy). In the second part the participants were asked to draw the figure again from recall 3 minutes after the first part was conducted (RCFT recall). Both parts were scored on a scale from 0 to 36 points based on features of the complex line drawing according to standard scoring criteria. The RCFT has been shown to measure visuospatial abilities, non-verbal memory, and planning (Meyers & Meyers, 1995).

Symbol Digit Modalities Test (SDMT)

Participants were given a response key with geometric symbols corresponding to different numbers and asked to substitute as many geometric symbols with numbers as possible in 90 seconds. The total score represents the number of correct substitutions. The test is a measure of processing speed, attention and working memory (Smith, 1982).

Modified Stroop Interference Test

The test consisted of 3 stimuli cards with four different colors (blue, black, red, green) presented in an alternating order 10 times on each card. The first stimuli card was a word-page with names of the colors. The second stimuli card was a color-page with the colors shown in boxes. The last stimuli card was a word-color page with incongruent conditions where the colors were written in non-matching colored ink. The task of the participants was to read the color names, name the colored boxes or name the color of the ink, respectively for the three stimuli cards, as quickly as possible without any errors. The reaction times represent the completion in seconds for the incongruent word-color card task, while the scores from the first two stimuli cards are not reported. The modified Stroop interference test examines executive functions, attention and processing speed (Klein et al., 1997).

Hearing Tests

All audiological assessments were administered by audiologists and performed in a sound-proof room. Prior to measuring the audiograms, all participants were examined with otoscopy and tympanometry to evaluate the condition of the external auditory canal and middle ear, and to rule out any conductive pathology or abnormalities that could affect hearing threshold measurements. If needed, the ear canals were cleansed for cerumen.

Audiometry

Pure-tone audiometry in the frequency range from 250 Hz to 8 kHz was measured using a standard clinical audiometer (model AS216, Interacoustics A/S, Middlefart, Denmark) and either DD450 headphones (RadioEar, Middelfart, Denmark) or IP30 insert earphones (RadioEar, Middelfart, Denmark). Bone conduction was performed, when necessary, with appropriate masking. The pure-tone average (PTA) was calculated for each participant based on the hearing threshold levels for the 500 Hz, 1 kHz, 2 kHz, and 4 kHz frequencies for the better hearing ear (BHE). The PTA BHE was used to classify the participants into HL categories based on the World Health Organization's grading system with the following categories: normal hearing below 20 dB, mild HL between 20 and 34 dB, moderate HL between 35 and 49 dB, moderately severe HL between 50 and 64 dB and severe HL between 65 and 79 dB (Humes, 2019; World Health Organization, 2021)

Sensory and Metabolic Hearing Estimates

Estimates of the sensory and metabolic components of HL were derived using a mathematical modeling approach developed by Vaden et al. (2022). The model fits individual audiogram data to a linear combination of canonical audiogram shapes that represent prototypical patterns of sensory and metabolic HL. These canonical shapes were defined based on average audiograms from individuals with well-characterized forms of each HL type. For each participant, the model provided an estimate (in dB) of the extent to which sensory and metabolic components contributed to the observed audiogram, based on the audiogram of the BHE. Following the suggested protocol, an error value was also calculated as the root mean square error (RMSE) between the modeled and actual audiogram, and audiograms with an RMSE greater than 15 dB were excluded from further analysis. Five participants (four healthy older adults and 1 MCI patient) were excluded from the analyses because their estimation error was above 15 dB.

Statistical Analysis

All analyses were performed in R Statistical language (R Core Team, 2022, Version 4.3.2 (2023-10-31)), with the following R-packages: Tidyverse (v. 2.0.0, (Wickham et al., 2019)) & ppcor (v. 1.1, (Kim, 2015)). Data were initially assessed for normality using the Shapiro–Wilk test. Due to non-normal distributions, non-parametric Spearman rank correlations were conducted to examine bivariate associations between HL components and cognitive test scores. Multiple comparisons were controlled using the false discovery rate (FDR) method to reduce the risk of Type I errors. In addition, to control for potential confounding effects of demographic variables, partial Spearman correlations were computed, adjusting for age and education where appropriate.

Finally, to examine whether the associations between HL components and cognitive performance differed between MCI patients and normal hearing older adults, linear regression analyses including interaction terms between HL component and group were conducted, with age included as a covariate.

Results

Demographics of Study Population

Table 1 shows information about the participant's sex, age, and hearing severity. The combined study population consisted of 55 female and 45 male participants with an age range from 55 to 91 years. The average age was 74.3 years. Most participants were normal hearing (n = 33) or had mild HL (n = 40). Approximately 30% of the participants had moderate to severe HL (n = 27), while 36 participants reported that they were HA users. Demographics of the MCI group compared to the cognitively healthy older adults’ group has been reported in an earlier study, where we show that the two groups are not significantly different in terms of sex, age, hearing severity, nor self-reported use of a HA (Hendel et al., 2026).

Table 1.

Summary of Demographic Characteristics of Participants.

Demographic Characteristic	Value
Sex
Female	55
Male	45
Age
Range, years	55–91
Mean (SD), years	74.3 (7.1)
Hearing severity
Normal	33
Mild	40
Moderate	16
Moderately severe	10
Severe	1
Hearing aid
Self-reported number of users	36

Metabolic and Sensory Hearing Loss Components

Figure 1A shows the average audiograms for the MCI group (blue) and the healthy older adults’ group (gray). Figure 1B shows the metabolic and sensory component of the HL for each participant. 33 had a metabolic HL, while 14 had a sensory HL. The remaining 48 had a combined sensory and metabolic HL. No statistically significant differences were found between the healthy older adult group and the MCI group when comparing both the estimates of the sensory and metabolic components (Figure 1C, sensory: W = 1025.5, P = .61, metabolic: W = 950, P = .28) and the average estimation errors (Figure 1D). Because there were no significant differences between the data of the two groups, the healthy older adults and the MCI groups were combined into one dataset in the next part of the analyses.

Figure 1.

Metabolic and sensory hearing loss (HL) components for two participant groups: healthy older adults and patients with mild cognitive impairment (MCI). (A) Average audiograms from the healthy older adults’ group (grey) and the MCI group (blue). (B) Scatter plot of the sensory and metabolic components of HL for each individual participant. (C) Group average estimates of the metabolic and sensory components for the healthy older adults and MCI groups. (D) Estimation errors of the fit between the audiometry data and the model templates. The grey circles are the individual estimation errors for the healthy older adults, while the blue circles are the individual estimation errors for the MCI group. The black diamonds represent the group averages, while the red line is the error estimation cut-off (15 dB).

Association Between Cognitive Tests and PTA

Table 2 presents initial correlation analyses between each of the cognitive tests and PTA BHE. PTA BHE was weakly and negatively correlated with LM immediate recall performance (r = −0.25, P < .05, Table 2, Figure 2A), and with LM delayed recall performance (ρ = −0.22, P < .05, Table 2). There was a moderate negative correlation between SDMT performance and PTA BHE (ρ = −0.48, P < .001, Table 2, Figure 2A), and a weak positive correlation with modified Stroop interference performance (ρ = 0.30, P < .01, Table 2, Figure 3A). No significant correlation was found between PTA BHE and the fluency and RCFT tests (Table 2). To assess the robustness of these associations, additional analyses controlling for age were performed; in these partial correlations, none of the associations between the PTA BHE and the cognitive tests remained statistically significant (all P > .10). The cognitive domains with significant correlations with PTA BHE in the unadjusted analyses were further analyzed below.

Figure 2.

Scatter plots of hearing thresholds as a function of Symbol Digit Modalities Test (SDMT) scores. (A) Relationship between PTA BHE and SDMT scores. (B) Relationship between the metabolic estimate and SDMT scores. (C) Relationship between the sensory estimate and SDMT scores.

Figure 3.

Scatter plots of hearing thresholds as a function of the modified Stroop interference test score. (A) Relationship between PTA BHE and results from the modified Stroop interference test. (B) Relationship between the metabolic estimate and results from the modified Stroop interference test. (C) Relationship between the sensory estimate and results from the modified Stroop interference test.

Table 2.

Correlations Between Pure Tone Averages (PTAs) of the Better Hearing ear (BHE) and Cognitive Test Performances.

Cognitive Test	P-Value	Spearman rho
Fluency, Animals	.092	−0.190
Fluency, FAS Total	.16	−0.157
LM Immediate Recall	.029^*	−0.254
LM Delayed Recall	.044^*	−0.219
RCFT Copy	.76	−0.031
RCFT Recall	.38	−0.098
SDMT	<.001^***	−0.476
Modified Stroop Interference Test	<.01^**	0.303

P-Values are shown after multiple comparisons correction.

P-Values < .05.

P-Values < .01.

***

P-Values < .001.

Association Between Cognitive Tests and Metabolic and Sensory HL

No significant association was found between the metabolic component and the LM immediate recall scores (ρ = −0.17, FDR-adjusted P = .19), nor between the sensory component and the LM immediate recall scores (ρ = −0.02, FDR-adjusted P = .83). Similarly, no significant association was found between the metabolic and sensory components and the LM delayed recall scores (metabolic: ρ = −0.18, FDR-adjusted P = .17, sensory: ρ = 0.01, FDR-adjusted P = .91).

A moderate negative correlation was found between the metabolic component and the SDMT score (ρ = −0.40, FDR-adjusted P < .001, Figure 2B) indicating that increased metabolic HL is associated with a poorer SDMT score. No significant association was found between the sensory component and the SDMT scores (ρ = 0.14, FDR-adjusted P = .17, Figure 2C). To examine whether the observed association between the SDMT performance and the metabolic HL component remained after accounting for age, a partial Spearman correlation was conducted. A statistically significant, negative weak correlation was found (ρ = −0.23, P = .024), indicating that worse metabolic HL is associated with worse SDMT performance even after adjusting for age.

A weak positive correlation was found between the metabolic component and the modified Stroop interference test score (ρ = 0.31, FDR-adjusted P < .01, Figure 3B) suggesting that larger metabolic HL is associated with worse Stroop test performance. No significant correlation was found between the sensory component and the Stroop test scores (ρ = −0.06, FDR-adjusted P = .55, Figure 3C). The correlation between the metabolic HL component and the Stroop interference test performance was no longer statistically significant when controlling for age (ρ = 0.11, P = .31).

Comparisons of MCI and Normal Hearing Older Adults

To examine whether the associations between cognitive performance and the metabolic and sensory HL components differed between participants with MCI and normal hearing older adults, additional linear regression analyses including interaction terms between hearing component and group were conducted, adjusting for age. No statistically significant interaction effects were observed for any cognitive outcome (all P > .10), indicating that the associations between HL components and cognitive performance did not differ between groups. Thus, the observed relationships were comparable in MCI and normal hearing older adults.

Discussion

The present study investigated the relationship between age-related HL and cognitive functioning in older adults, with particular attention to the differential impact of sensory and metabolic components of HL. Using pure-tone audiometry and a comprehensive cognitive test battery, significant associations were found between hearing thresholds and cognitive performance, particularly in the domains of processing speed and executive functions. Crucially, these associations were primarily driven by the metabolic component of HL. However, when controlling for age, the association between the metabolic component and executive functions was no longer significant, suggesting that age may act as a confounding variable in this relationship. No significant associations were observed between the sensory component and any cognitive domains.

These findings are consistent with a growing body of literature suggesting that HL, particularly age-related HL, is related to poorer cognitive performance (Cantuaria et al., 2024; Deal et al., 2017; Gallacher et al., 2012; Gurgel et al., 2014; Lin et al., 2011; Loughrey et al., 2018; Powell et al., 2022; Strutt et al., 2022). In the current study, episodic memory, executive function and processing speed were all associated with PTA when age was not included as a covariate. Several previous studies have reported similar results for episodic memory (Matthews et al., 2023; Rönnberg et al., 2011, 2014) and executive functions (Brewster et al., 2021). Notably, these earlier studies included much larger sample sizes, suggesting that the lack of a significant correlation with executive function in the present study—more specifically on the Stroop test—may be due to limited statistical power.

Nevertheless, the correlation between SDMT and metabolic estimate remained significant even after controlling for age, indicating that age alone does not fully account for the association between hearing thresholds and decreased processing speed. This finding is consistent with a recent study by Nishiyama et al. (2025), which also reported a negative correlation between hearing thresholds and SDMT scores in older adults who did not use HAs. Importantly, while most prior studies have treated HL as a unitary phenomenon, the present study provides an important nuance, namely that sensory and metabolic HL may affect cognition differently.

This study did not find any significant correlations between PTA and fluency or RCFT performance, suggesting that hearing thresholds were not associated with visuoconstructive abilities, non-verbal memory, or verbal fluency in this cohort. While large meta-analyses report small but statistically significant associations between age-related HL and performance across multiple cognitive domains, including visuospatial ability and verbal fluency measures, the strength and consistency of these associations vary across study designs and specific tasks, with some longitudinal data showing weaker or non-significant links for verbal fluency specifically (Loughrey et al., 2017).

Taken together, the present findings suggest a differentiated pattern in which hearing thresholds —particularly the metabolic component of HL—were most consistently related to processing speed, whereas associations with executive functions and visuoconstruction were less robust or absent after adjustment for age. This may indicate that age-related HL in this cohort is more closely linked to slowing of cognitive processing than to non-auditory visuospatial abilities. However, given the modest sample size and the small effect sizes typically reported in the literature, these domain-specific interpretations should be considered with caution.

It is important to also highlight a difference to most prior studies, namely that this study included two subpopulations where one consisted of healthy older adults from the general public, while the other consisted of patients diagnosed with MCI from the memory clinic at Rigshospitalet, Denmark. The MCI patients were generally performing worse than the healthy older adults in the cognitive tests (Hendel et al., 2026) which could lead to stronger correlations in the present study than previous studies only including healthy older adults. Importantly though, analyses examining interactions between group (MCI vs. healthy older adults) and HL components did not reveal any statistically significant interaction effects. This indicates that the associations between metabolic HL and cognitive performance did not differ significantly between the two subpopulations. Thus, the observed relationship between metabolic HL and processing speed appears to be consistent across both cognitively healthy older adults and individuals with MCI.

The distinction between sensory and metabolic HL may have important implications for understanding the biological mechanisms underlying cognitive decline. For example, there is evidence that genes involved with metabolic HL have also previously been implicated in amyotrophic lateral sclerosis and fronto-temporal dementia (Ahmed et al., 2025), suggesting that a shared pathological pathway may be involved in both metabolic HL and cognitive decline. The mechanisms of the pathological pathway between hearing and cognitive decline are not yet understood, but several theories exist, including vascular pathologies, mitochondrial dysfunction, or systemic inflammation (Kostrikov et al., 2025).The idea of a shared vascular pathology is supported by the increased risk of HL that is observed in individuals with hypertension, as well as by the fact that hypertension has been found to increase the risk of HL after occupational noise exposure (Zhang et al., 2023) and furthermore, hypertension is named as a risk factor for dementia (Livingston et al., 2024). The current study was not designed to test the hypothesis of a shared vascular pathology and therefore did not include any objective vascular measures. Instead, a simple post-hoc analysis was conducted to explore any possible links between self-reported hypertension and metabolic HL. However, the participants with hypertension did not show a higher degree of metabolic HL, nor did the MCI group have a significantly higher occurrence of hypertension compared to the older healthy adults (data not shown). While this result is inconsistent with the idea that a vascular pathology could explain the link between metabolic HL and SDMT, it is worth noting that this is based on a self-report of hypertension, and no objective measures were included. Furthermore, it is worth noting that the self-report did not evaluate whether the participants were actively treating their hypertension. It is possible that no link was found between the hypertension and metabolic HL because pharmacological treatment is highly effective and that most of the participants were treating their hypertension. Future studies would benefit from including objective measures of vascular pathologies and treatment histories to more thoroughly investigate the hypothesis of a shared pathology. Although the present findings may reflect a shared pathological mechanism, it is unlikely that a single pathway fully accounts for the association between HL and cognitive decline. Instead, both HL and cognitive impairment are likely shaped by multiple processes, with considerable inter-individual variability in their relative contributions.

Given the central role of the metabolic and sensory HL estimates in interpreting the findings in the current study, it is important to consider the validity and limitations of the modeling approach used to derive them. Only five participants had to be removed from the dataset due to an estimate error above 15 dB, and the average estimate error was 7.75 dB. Furthermore, larger metabolic estimates were associated with increased age, which is consistent with findings in earlier studies (Vaden et al., 2016, 2022). Additionally, males have previously been shown to have signiﬁcantly higher sensory estimates compared to females (Dubno et al., 2013; Vaden et al., 2016, 2022)—a finding this study also replicates (Supplementary Figure S1). Given the low estimate error and the similarities in key trends in the data compared to previous literature, the model appeared to be well-suited for characterizing the audiograms in the present study. However, not all research agrees with the fundamental assumption that metabolic HL in the form of strial degeneration can explain the audiometric patterns in aging that underlies the employed model (Dubno et al., 2013; Vaden et al., 2022). Kaur et al. (2023) reported that while the overall shape of the audiogram is not a strong predictor of SV atrophy, elevated low-frequency thresholds show a modest correlation with SV degeneration. Furthermore, Wu et al. (2020) argued that strial degeneration cannot be identified using the audiogram because at the time of strial tissue loss, the hair cell death is so extensive that the additional strial degeneration does not further decrease the pure-tone thresholds. Instead, it was suggested that cochlear synaptopathy and neural degeneration were the main components of age-related auditory decline (Wu et al., 2020). This would suggest that strial degeneration may not play a significant role in age-related HL, and therefore, a shared pathological mechanism between cognitive decline and metabolic HL may be less likely. This alternative perspective highlights the need for caution when interpreting HL subtypes based solely on audiometric data and underscores the importance of further validating these components with physiological data in future work. Furthermore, it is important to acknowledge that the available human temporal bone collections on which the model is based may be subject to sample bias, over-representing individuals with clinical otologic conditions. This underscores the importance of validating the model using animal data with known pathology. Vaden et al. (2022) performed such validation using gerbil groups with verified strial and outer hair cell pathology, although further animal work would strengthen confidence in these interpretations. Additionally, correlations between PTA scores and the metabolic and sensory components showed that the metabolic component is more correlated with the PTA scores than the sensory component. This can be explained by the fact that the metabolic profile has a higher degree of HL in the low frequencies that are included in the calculated PTA, while the sensory profile has a higher degree of HL in the high frequencies that are not included in the calculation of PTA (Vaden et al., 2022). Thus, the fact that processing speed is more correlated to metabolic HL than sensory HL may be confounded by the fact that the metabolic component is itself more correlated to PTA. We are unable to disentangle these aspects in the current study design. Additionally, a limitation of the current study is that data was not collected on participants’ lifetime noise exposure, which is known to influence hearing status. Thus, future studies should include assessments of noise exposure as well to better tease apart the interplay between these aspects.

In conclusion, this study suggests that metabolic, but not sensory, components of age-related HL are linked to decline in processing speed and executive functioning. While age is a confounding factor in the association between metabolic HL and executive functions, the association between metabolic HL and processing speed remained significant after correcting for age. These results highlight the importance of considering the etiology of HL when investigating its cognitive consequences. They also support emerging hypotheses that suggest a shared pathophysiology between age-related HL and cognitive decline. However, this study used an indirect proxy for assessing an underlying pathology to HL; thus, future research validating the model with physiological data will be critical for further understanding the mechanisms linking metabolic HL to cognitive decline.

Supplemental Material

sj-docx-1-tia-10.1177_23312165261447076 - Supplemental material for Association Between Age-Related Sensory and Metabolic Hearing Loss and Cognitive Performance

Supplemental material, sj-docx-1-tia-10.1177_23312165261447076 for Association Between Age-Related Sensory and Metabolic Hearing Loss and Cognitive Performance by Mie L. Jørgensen, Rebecca K. Hendel, Asmus Vogel and Abigail Anne Kressner in Trends in Hearing

Footnotes

Acknowledgments

The authors would like to thank the clinical neuropsychologists from the Memory Clinic at the Danish Dementia Research Centre, Rigshospitalet, Denmark for their help with recruitment and clinical assessment of cognition of the MCI patients. Furthermore, we would like to thank the audiologists Amal Abdulqadir Ali, Sagal Rådberg Nagbøl, and Sofie Kobberø Bundgaard from the Copenhagen Hearing and Balance Center at Rigshospitalet, Denmark for their audiological assessment. Finally, we would like to thank Emilie Najbjerg Birkebæk and Lea Virenfeldt Damgaard for their help with the cognitive assessment of the healthy older adults.

ORCID iDs

Mie L. Jørgensen

Rebecca K. Hendel

Asmus Vogel

Abigail Anne Kressner

Ethical Considerations

The study was approved by the Ethics Committee of the Capital Region of Denmark (F-23054924) and written informed consent was obtained from all participants before participation in the study. The study was conducted in accordance with the Declaration of Helsinki.

Author Contributions

MLJ, RKH, AV, and AAK conceived and designed the study. RKH was responsible for the recruitment and collection of data. MLJ and AAK planned and performed the analysis. MLJ, AAK, AV, and RKH interpreted the statistical analyses of the data. MLJ prepared the first draft of the manuscript, and RKH, AV and AAK contributed to the revision process. The submitted version of the article has been approved by all authors.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was funded by the GN Foundation. The foundation was not involved in the study design, conduction, or interpretation.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data Availability

The dataset is only available from the corresponding author upon reasonable request due to privacy and legal restrictions.

Supplemental Material

Supplemental material for this article is available online.

References

Ahmed

Kenneth I Vaden

Leitao

Dubno

J. R.

Drögemöller

B. I.

(2025). Large-scale audiometric phenotyping identifies distinct genes and pathways involved in hearing loss subtypes. MedRxiv, 2025.01.14.24318673. https://doi.org/10.1101/2025.01.14.24318673

Alattar

A. A.

Bergstrom

Laughlin

G. A.

Kritz-Silverstein

Richard

E. L.

Reas

E. T.

Harris

J. P.

Barrett-Connor

McEvoy

L. K.

(2020). Hearing impairment and cognitive decline in older, community-dwelling adults. The Journals of Gerontology. Series A, Biological Sciences and Medical Sciences, 75(3), 567–573. https://doi.org/10.1093/GERONA/GLZ035

Bovee

Klump

G. M.

Köppl

Pyott

S. J.

(2024). The stria vascularis: Renewed attention on a key player in age-related hearing loss. International Journal of Molecular Sciences, 25(10). https://doi.org/10.3390/IJMS25105391

Brewster

K. K.

M. C.

Wall

M. M.

Brown

P. J.

Zilcha-Mano

Roose

S. P.

Stein

Golub

J. S.

Rutherford

B. R.

(2021). Age-related hearing loss, neuropsychological performance, and incident dementia in older adults. Journal of Alzheimer’s Disease: JAD, 80(2), 855. https://doi.org/10.3233/JAD-200908

Brito

D. V. C.

Esteves

Rajado

A. T.

Silva

Andrade

Apolónio

Calado

Faleiro

Matos

Marques

Marreiros

Nzwalo

Pais

Palmeirim

Roberto

Simão

Joaquim

Miranda

Pêgas

Nóbrega

(2023). Assessing cognitive decline in the aging brain: Lessons from rodent and human studies. Npj Aging, 9(1). https://doi.org/10.1038/S41514-023-00120-6

Cantuaria

M. L.

Pedersen

E. R.

Waldorff

F. B.

Wermuth

Pedersen

K. M.

Poulsen

A. H.

Raaschou-Nielsen

Sørensen

Schmidt

J. H.

(2024). Hearing loss, hearing aid use, and risk of dementia in older adults. JAMA Otolaryngology—Head & Neck Surgery, 150(2), 157–164. https://doi.org/10.1001/JAMAOTO.2023.3509

Chan

D. K.

Hudspeth

A. J.

(2005). Ca2+current-driven nonlinear amplification by the mammalian cochlea in vitro. Nature Neuroscience, 8(2), 149. https://doi.org/10.1038/NN1385

Deal

J. A.

Betz

Yaffe

Harris

Purchase-Helzner

Satterfield

Pratt

Govil

Simonsick

E. M.

Lin

F. R.

(2017). Hearing impairment and incident dementia and cognitive decline in older adults: The health ABC study. The Journals of Gerontology: Series A, 72(5), 703–709. https://doi.org/10.1093/GERONA/GLW069

Dubno

J. R.

Eckert

M. A.

Lee

F. S.

Matthews

L. J.

Schmiedt

R. A.

(2013). Classifying human audiometric phenotypes of age-related hearing loss from animal models. Journal of the Association for Research in Otolaryngology: JARO, 14(5), 687–701. https://doi.org/10.1007/S10162-013-0396-X

10.

Folstein

M. F.

Folstein

S. E.

Mchugh

P. R.

(1975). ‘Mini-mental state’ a practical method for grading the cognitive state of patients for the clinician*. Journal of Psychiatric Research, 12(3), 189–198. https://doi.org/10.1016/0022-3956(75)90026-6

11.

Gallacher

Ilubaera

Ben-Shlomo

Bayer

Fish

Babisch

Elwood

(2012). Auditory threshold, phonologic demand, and incident dementia. Neurology, 79(15), 1583–1590. https://doi.org/10.1212/WNL.0B013E31826E263D

12.

Golub

J. S.

Luchsinger

J. A.

Manly

J. J.

Stern

Mayeux

Schupf

(2017). Observed hearing loss and incident dementia in a multiethnic cohort. Journal of the American Geriatrics Society, 65(8), 1691–1697. https://doi.org/10.7916/Z53V-ND38

13.

Gurgel

R. K.

Ward

P. D.

Schwartz

Norton

M. C.

Foster

N. L.

Tschanz

J. T.

(2014). Relationship of hearing loss and dementia: A prospective, population-based study. Otology & Neurotology : Official Publication of the American Otological Society, American Neurotology Society [and] European Academy of Otology and Neurotology, 35(5), 775–781. https://doi.org/10.1097/MAO.0000000000000313

14.

Harada

C. N.

Natelson Love

M. C.

Triebel

K. L.

(2013). Normal cognitive aging. Clinics in Geriatric Medicine, 29(4), 737–752. https://doi.org/10.1016/j.cger.2013.07.002

15.

Hendel

R. K.

Kressner

A. A.

Jørgensen

M. L.

Dau

Waldemar

Vogel

(2026). A comparison of hearing abilities in memory clinic patients with mild cognitive impairment and cognitively intact older adults. Journal of Alzheimer’s Disease: JAD, 110(4), 1713–1722. https://doi.org/10.1177/13872877261427356

16.

Hughes

C. P.

Berg

Danziger

W. L.

Coben

L. A.

Martin

R. L.

(1982). A new clinical scale for the staging of dementia. The British Journal of Psychiatry, 140(6), 566–572. https://doi.org/10.1192/BJP.140.6.566

17.

Humes

L. E.

(2019). The World Health Organization’s hearing-impairment grading system: An evaluation for unaided communication in age-related hearing loss. International Journal of Audiology, 58(1), 12–20. https://doi.org/10.1080/14992027.2018.1518598

18.

Kaur

P.-Z.

O’malley

J. T.

Liberman

M. C.

(2023). Predicting atrophy of the cochlear stria vascularis from the shape of the threshold audiogram. https://doi.org/10.1523/JNEUROSCI.1138-23.2023

19.

Kim

(2015). Ppcor: An R package for a fast calculation to semi-partial correlation coefficients. Communications for Statistical Applications and Methods, 22(6), 665. https://doi.org/10.5351/CSAM.2015.22.6.665

20.

Klein

Ponds

R. W. H. M.

Houx

P. J.

Jolles

(1997). Effect of test duration on age-related differences in stroop interference. Journal of Clinical and Experimental Neuropsychology, 19(1), 77–82. https://doi.org/10.1080/01688639708403838

21.

Kostrikov

Hjortkjaer

Dau

(2025). Vascular pathology in age-related hearing loss. Ageing and Longevity, 6(1), 69–72. https://doi.org/10.47855/JAL9020-2025-1-9

22.

Lezak

M. D.

Howieson

D. B.

Bigler

E. D.

Tranel

(2004). Verbal functions and language skills. In Neuropsychological assessment (4th ed., pp. 501–530). Oxford University Press.

23.

Lin

F. R.

Metter

E. J.

O’Brien

R. J.

Resnick

S. M.

Zonderman

A. B.

Ferrucci

(2011). Hearing loss and incident dementia. Archives of Neurology, 68(2), 214–220. https://doi.org/10.1001/ARCHNEUROL.2010.362

24.

Lin

F. R.

Pike

J. R.

Albert

M. S.

Arnold

Burgard

Chisolm

Couper

Deal

J. A.

Goman

A. M.

Glynn

N. W.

Gmelin

Gravens-Mueller

Hayden

K. M.

Huang

A. R.

Knopman

Mitchell

C. M.

Mosley

Pankow

J. S.

Reed

N. S.

Coresh

(2023). Hearing intervention versus health education control to reduce cognitive decline in older adults with hearing loss in the USA (ACHIEVE): A multicentre, randomised controlled trial. The Lancet, 402(10404), 786–797. https://doi.org/10.1016/S0140-6736(23)01406-X

25.

Livingston

Huntley

Liu

K. Y.

Costafreda

S. G.

Selbaek

Alladi

Ames

Banerjee

Burns

Brayne

Fox

N. C.

Ferri

C. P.

Gitlin

L. N.

Howard

Kales

H. C.

Kivimäki

Larson

E. B.

Nakasujja

Rockwood

Mukadam

(2024). Dementia prevention, intervention, and care: 2024 report of the Lancet Standing Commission. The Lancet, 404(10452), 572–628. https://doi.org/10.1016/S0140-6736(24)01296-0

26.

Livingston

Huntley

Sommerlad

Ames

Ballard

Banerjee

Brayne

Burns

Cohen-Mansfield

Cooper

Costafreda

S. G.

Dias

Fox

Gitlin

L. N.

Howard

Kales

H. C.

Kivimäki

Larson

E. B.

Ogunniyi

Mukadam

(2020). Dementia prevention, intervention, and care: 2020 report of the Lancet Commission. The Lancet, 396(10248), 413–446. https://doi.org/10.1016/S0140-6736(20)30367-6

27.

Loughrey

D. G.

Kelly

M. E.

Kelley

G. A.

Brennan

Lawlor

B. A.

(2017). Association of age-related hearing loss with cognitive function, cognitive impairment, and dementia: A systematic review and meta-analysis. JAMA Otolaryngology—Head & Neck Surgery, 144(2), 115. https://doi.org/10.1001/JAMAOTO.2017.2513

28.

Loughrey

D. G.

Kelly

M. E.

Kelley

G. A.

Brennan

Lawlor

B. A.

(2018). Association of age-related hearing loss with cognitive function, cognitive impairment, and dementia: A systematic review and meta-analysis. JAMA Otolaryngology–Head & Neck Surgery, 144(2), 115–126. https://doi.org/10.1001/JAMAOTO.2017.2513

29.

Matthews

Dawes

Elliot

Pendleton

Tampubolon

Maharani

(2023). Trajectories of self-reported hearing and their associations with cognition: Evidence from the United Kingdom and United States of America. Age and Ageing, 52(2), 1–9. https://doi.org/10.1093/AGEING/AFAD017

30.

Meyers

J. E.

Meyers

K. R.

(1995). Rey complex figure test and recognition trial. PAR.

31.

Nichols

Steinmetz

J. D.

Vollset

S. E.

Fukutaki

Chalek

Abd-Allah

Abdoli

Abualhasan

Abu-Gharbieh

Akram

T. T.

Al Hamad

Alahdab

Alanezi

F. M.

Alipour

Almustanyir

Amu

Ansari

Arabloo

Ashraf

Vos

(2022). Estimation of the global prevalence of dementia in 2019 and forecasted prevalence in 2050: An analysis for the global burden of disease study 2019. The Lancet Public Health, 7(2), e105–e125. https://doi.org/10.1016/S2468-2667(21)00249-8/ATTACHMENT/60E03FD1-38B2-4B40-A91D-9AFDDA22B45E/MMC1.PDF

32.

Nishiyama

Kimizuka

Kataoka

Tazoe

Sato

Hosoya

Shimanuki

M. N.

Wakabayashi

Ueno

Ozawa

Oishi

(2025). Relationship between hearing thresholds and cognitive function in hearing aid non-users and long-term users post-midlife. NPJ Aging, 11(1), 14. https://doi.org/10.1038/S41514-025-00203-6

33.

Petersen

R. C.

(2016). Mild cognitive impairment. Continuum, 22(2), 404–418. https://doi.org/10.1212/CON.0000000000000313

34.

Powell

D. S.

E. S.

Reed

N. S.

Lin

F. R.

Deal

J. A.

(2022). Hearing loss and cognition: What we know and where we need to go. Frontiers in Aging Neuroscience, 13, 769405. https://doi.org/10.3389/FNAGI.2021.769405

35.

Rönnberg

Danielsson

Rudner

Arlinger

Sternäng

Wahlin

Nilssond

L. G.

(2011). Hearing loss is negatively related to episodic and semantic long-term memory but not to short-term memory. Journal of Speech, Language, and Hearing Research, 54(2), 705–726. https://doi.org/10.1044/1092-4388(2010/09-0088)

36.

Rönnberg

Hygge

Keidser

Rudner

(2014). The effect of functional hearing loss and age on long- and short-term visuospatial memory: Evidence from the UK biobank resource. Frontiers in Aging Neuroscience, 6(DEC). https://doi.org/10.3389/FNAGI.2014.00326

37.

Schuknecht

H. F.

Gacek

M. R.

(1993). Cochlear pathology in presbycusis. Annals of Otology, Rhinology & Laryngology, 102(1 II), 1–16. https://doi.org/10.1177/00034894931020S101

38.

Shah

Salek

M. S.

Ali

F. M.

Nixon

S. J.

Otwombe

Ingram

J. R.

Finlay

A. Y.

(2024). Dementia and its profound impact on family members and partners: A large UK cross-sectional study. Alzheimer Disease and Associated Disorders, 38(4), 338. https://doi.org/10.1097/WAD.0000000000000647

39.

Smith

(1982). Symbol digit modalities test. In Manual. Western Psychological Services.

40.

Strutt

P. A.

Barnier

A. J.

Savage

Picard

Kochan

N. A.

Sachdev

Draper

Brodaty

(2022). Hearing loss, cognition, and risk of neurocognitive disorder: Evidence from a longitudinal cohort study of older adult Australians. Aging, Neuropsychology, and Cognition, 29(1), 121–138. https://doi.org/10.1080/13825585.2020.1857328

41.

Sugawara

Sasaki

Yasui-Furukori

Kakehata

Umeda

Namba

Nakaji

Shinkawa

Kaneko

(2011). Hearing impairment and cognitive function among a community-dwelling population in Japan. Annals of General Psychiatry, 10(1), 27. https://doi.org/10.1186/1744-859X-10-27

42.

Sweller

(2011). Cognitive load theory. Psychology of Learning and Motivation—Advances in Research and Theory, 55, 37–76. https://doi.org/10.1016/B978-0-12-387691-1.00002-8

43.

Vaden

K. I.

Eckert

M. A.

Matthews

L. J.

Schmiedt

R. A.

Dubno

J. R.

(2022). Metabolic and sensory components of age-related hearing loss. JARO—Journal of the Association for Research in Otolaryngology, 23(2), 253–272. https://doi.org/10.1007/S10162-021-00826-Y

44.

Vaden

K. I.

Matthews

L. J.

Dubno

J. R.

(2018). Transient-evoked otoacoustic emissions reflect audiometric patterns of age-related hearing loss. Trends in Hearing, 22, 2331216518797848. https://doi.org/10.1177/2331216518797848

45.

Vaden

K. I.

Matthews

L. J.

Eckert

M. A.

Dubno

J. R.

(2016). Longitudinal changes in audiometric phenotypes of age-related hearing loss. JARO: Journal of the Association for Research in Otolaryngology, 18(2), 371. https://doi.org/10.1007/S10162-016-0596-2

46.

Wechsler

(1997). Wechsler memory scale (3rd ed.).

47.

Wickham

Averick

Bryan

Chang

D’

Mcgowan

François

Grolemund

Hayes

Henry

Hester

Kuhn

Lin Pedersen

Miller

Bache

S. M.

Müller

Ooms

Robinson

Seidel

D. P.

Yutani

(2019). Welcome to the tidyverse. Journal of Open Source Software, 4(43), 1686. https://doi.org/10.21105/JOSS.01686

48.

Wimo

Seeher

Cataldi

Cyhlarova

Dielemann

J. L.

Frisell

Guerchet

Jönsson

Malaha

A. K.

Nichols

Pedroza

Prince

Knapp

Dua

(2023). The worldwide costs of dementia in 2019. Alzheimer’s and Dementia, 19(7), 2865–2873. https://doi.org/10.1002/ALZ.12901

49.

Winblad

Palmer

Kivipelto

Jelic

Fratiglioni

Wahlund

L. O.

Nordberg

Bäckman

Albert

Almkvist

Arai

Basun

Blennow

De Leon

Decarli

Erkinjuntti

Giacobini

Graff

Hardy

Petersen

R. C.

(2004). Mild cognitive impairment – beyond controversies, towards a consensus: Report of the International Working Group on mild cognitive impairment. Journal of Internal Medicine, 256(3), 240–246. https://doi.org/10.1111/J.1365-2796.2004.01380.X

50.

World Health Organization. (2021). Challenges facing ear and hearing care. World Report on Hearing, 139–198. https://apps.who.int/iris/rest/bitstreams/1334317/retrieve

51.

P. Z.

O’Malley

J. T.

de Gruttola

Charles Liberman

(2020). Age-related hearing loss is dominated by damage to inner ear sensory cells, not the cellular battery that powers them. The Journal of Neuroscience, 40(33), 6357. https://doi.org/10.1523/JNEUROSCI.0937-20.2020

52.

Zhang

H. D.

Yuan

Jin

Deng

H. X.

(2023). The combined effect of elevated blood pressure and occupational noise exposure on bilateral high-frequency hearing loss: Evidence from a large sample cross-sectional study. Journal of Occupational and Environmental Medicine, 65(4), e219. https://doi.org/10.1097/JOM.0000000000002783

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.38 MB

0.00 MB