Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2026 Sep 26.
Published in final edited form as: Neurobiol Dis. 2026 Jul 15;228:107535. doi: 10.1016/j.nbd.2026.107535

Neurovascular-metabolic coupling links hearing loss to cognitive impairment: Evidence from GABA and cerebral blood flow in presbycusis

Yao Wang a,b,c, Shuya Wang a, Wen Ma d, Wenqing Li a, Weilong Fu a, Fuxin Ren e, Bing Liu e, Richard AE Edden f,g, Muwei Li h,i, Xinjuan Zhang e,*, Fei Gao e,*
PMCID: PMC13613338  NIHMSID: NIHMS2208318  PMID: 42456953

Abstract

Hearing loss is recognized as the leading modifiable risk factor for dementia. However, the neurobiological pathways bridging auditory dysfunction and cognitive decline remain poorly understood. We hypothesized that a neurovascular-metabolic cascade, involving neurotransmitter imbalances and perfusion deficits, underlies this link. We recruited 124 patients with presbycusis and 91 age-matched health controls. Auditory GABA and glutamate levels were quantified using macromolecule-suppressed MEGA-PRESS, and cerebral blood flow (CBF) was quantified using pseudo-continuous arterial spin labeling. Participants underwent comprehensive neuropsychological assessments. Spatial correlation analysis was performed to align CBF alterations with established neurotransmitter maps. Presbycusis patients exhibited bilateral reductions in auditory GABA and glutamate, alongside widespread hypoperfusion across auditory, salience, and executive control networks. Mediation analysis demonstrated that right auditory GABA concentrations and regional CBF (specifically in the right superior parietal gyrus and left thalamus) sequentially mediated the association between hearing loss and cognitive deterioration. Furthermore, CBF alterations spatially co-localized with serotonergic, dopaminergic, and GABAergic systems, with these spatial couplings significantly correlating with cognitive performance. These findings delineate a systemic neurovascular-metabolic cascade linking hearing loss to cognitive impairment. By identifying GABAergic signaling and regional perfusion as key mediators, this study provides a multimodal framework for understanding sensory-cognitive aging and highlights potential biomarkers for early intervention in dementia.

Keywords: Neurovascular-metabolic, Hearing loss, Cognitive impairment, GABA, Cerebral blood flow

1. Introduction

Hearing impairment is currently regarded as the foremost modifiable contributor to dementia risk (Livingston et al., 2020). Long-standing, untreated hearing impairment can hasten the deterioration of cognitive faculties and heighten susceptibility to dementia (Lin and Albert, 2014; Loughrey et al., 2018). Presbycusis refers to the progressive, bilateral degeneration of sensorineural hearing associated with aging (Guan et al., 2022). It is characterized by decreased hearing sensitivity, impaired speech comprehension, and slowed central auditory processing, with particular susceptibility to background noise (Huang and Tang, 2010). Accumulating evidence has linked presbycusis to cognitive impairment in processing speed, executive function, and memory (Lin et al., 2011a; Lin et al., 2011b). However, the underlying neurometabolic mechanisms remain elucidated.

Structural alterations in auditory and cognitive brain regions have been documented in presbycusis using magnetic resonance imaging (MRI) (Chen et al., 2020; Delano et al., 2020). Atrophic changes in gray matter are evident in the primary auditory cortex and neighboring temporal lobe structures (Eckert et al., 2012; Peelle et al., 2011). Surface-based morphometry has further demonstrated volume loss in hubs within default mode network (DMN), specifically the bilateral precuneus and right posterior cingulate cortex (Ren et al., 2018). In presbycusis, resting-state fMRI has revealed a significant association between hippocampal-inferior parietal connectivity and performance on the Trail Making Test B (Chen et al., 2020). Evidence from multiple studies indicates that enhanced functional connectivity between the dorsolateral prefrontal cortex and dorsal auditory stream, and diminished connectivity linking the posterior cingulate to DMN regions (Ren et al., 2021). Current research on the neural mechanisms underlying sensory-cognitive interactions has shifted from focusing on isolated brain structural changes to investigating whole-brain functional network reorganization. For example, graph theory-based analyses have revealed that presbycusis not only induces functional alterations in the primary auditory cortex but also leads to topological abnormalities in both low-order and high-order functional networks, particularly within the default mode network and frontal regions (Xu et al., 2024). Collectively, these structural and functional changes suggest presbycusis affects distributed networks beyond the auditory cortex, potentially contributing to its associated cognitive deficits.

Cerebral blood flow (CBF) is an important physiological indicator of brain function. Individuals with sudden sensorineural hearing loss exhibit significantly reduced CBF in the middle frontal gyrus, which correlates with hearing severity (Chen et al., 2022). The whole-brain CBF in healthy elderly adults decreased significantly with age, which has been linked to cognitive decline in processing speed, working memory, and episodic recall (De Vis et al., 2018). A longitudinal arterial spin labeling (ASL) study showed that decreased CBF in the temporal and frontal lobes predicted memory decline in vascular cognitive impairment (van Dinther et al., 2023). One pseudo-continuous ASL (PCASL) study demonstrated reduced CBF in the right auditory cortex of presbycusis, associated with hearing loss (Ponticorvo et al., 2019). However, whether perfusion in other regions is similarly altered and how it relates to cognition remains unknown.

Glutamate (Glu) and γ-aminobutyric acid (GABA), the principal excitatory and inhibitory neurotransmitters in the central nervous system, are essential for maintaining homeostasis in central auditory processing (Bartos et al., 2007). GABA additionally contributes to neural plasticity and oscillatory synchronization, processes that are both essential for normal cognitive function (Lalwani et al., 2019; Sumner et al., 2010). One magnetic resonance spectroscopy (MRS) study has revealed reduced Glu in the auditory cortex of presbycusis (Profant et al., 2013), while MEGA-PRESS studies revealed decreased auditory GABA in presbycusis, correlated with hearing impairment and executive dysfunction (Gao et al., 2015; Li et al., 2023). GABA concentrations in the visual cortex exhibit a positive correlation with regional CBF (Donahue et al., 2010). In high-risk psychosis populations, dorsolateral prefrontal GABA correlated positively with hippocampal CBF and negatively with ventrolateral prefrontal CBF (Modinos et al., 2018). Furthermore, the sensory deprivation hypothesis posits that chronic reduction of auditory input results in cross-modal cortical reorganization. Su et al. (Su et al., 2024) demonstrated, using a combination of MRS and dynamic functional connectivity analyses, that downregulation of the GABAergic system and alterations in excitation/inhibition balance in the auditory cortex constitute the key neurochemical basis driving this reorganization. This neurochemical imbalance, along with alterations in network dynamics, jointly forms the micro- and macro-scale mechanisms underlying the reorganization of the “cognitive-auditory link”. Despite these findings, how GABA and Glu levels relate to CBF alterations in presbycusis remains largely unknown.

In addition, a study found that the hearing ability in healthy elderly people is significantly correlated with neurotransmitters related to the dopaminergic system, such as dopamine transporters (Qiu et al., 2024). Further, Dukart et al. (Dukart et al., 2021) used the JuSpace toolbox for cross-modal spatial analysis and found that drug-induced changes in CBF were highly correlated with their target distribution in the serotonin and dopaminergic systems, which provided methodological support for understanding the spatial association between neurotransmitter-induced CBF. Although multiple hypotheses have been proposed regarding the mechanisms linking presbycusis to cognitive decline, recent reviews emphasize that it is insufficient to consider auditory system pathology in isolation from the broader brain context. Paciello et al. (Paciello et al., 2026) systematically outlined how hearing loss may accelerate cognitive decline in Alzheimer’s disease models by exacerbating oxidative stress, neuroinflammation, and synaptic dysfunction, indicating that presbycusis and neurodegenerative disorders share similar molecular pathological substrates. However, there is no systematic study to explore the spatial correlation of neurotransmitter-CBF the process of cognitive decline caused by hearing loss in presbycusis patients.

Therefore, we used macromolecule-suppressed MEGA-PRESS and ASL to quantify auditory neurotransmitter levels and regional CBF in presbycusis and matched healthy controls. We aimed to: (1) examine group differences in GABA, Glu, and CBF levels; (2) assess how these neurobiological markers relate to auditory and cognitive performance; (3) test whether neurotransmitter concentrations and CBF act as mediators linking hearing loss to cognitive decline; (4) explore whether CBF alterations showed spatial co-localization with the distributions of key neurotransmitter systems. These findings will advance understanding of the neurochemical and vascular pathways underlying sensory-cognitive integration in the aging presbycusis.

2. Methods

2.1. Participants

This study included 124 individuals with presbycusis (72 males, 52 females; age range: 53–76 years) and 91 age-, sex-, and education-matched normal-hearing (NH) controls (48 males, 43 females; age range: 55–72 years). All participants were right-handed, Han Chinese, and native Mandarin speakers. Hearing thresholds were measured using the pure-tone average (PTA) across 0.5, 1, 2, and 4 kHz. Individuals in the presbycusis group met the criterion of PTA exceeding 25 dB HL in their better ear (Lin et al., 2011c). Exclusion criteria included: (1) asymmetrical or conductive hearing loss, Meniere’s disease, acoustic neuroma, tinnitus, or self-reported hyperacusis; (2) history of ototoxic drug use, ear surgery, head trauma, stroke, current or prior noise exposure, or use of hearing aids; (3) neurological or psychiatric disorders; (4) contraindications to MRI. Approval for this research was granted by the Institutional Review Board of Shandong University (approval number: 2016-KY-059), with all subjects signing informed consent forms before participation.

2.2. Auditory assessment

Pure-tone thresholds were obtained at octave frequencies from 0.125 to 8 kHz using a Madsen Electronics Midimate 622 device with TDH-39P headphones. Speech recognition thresholds (SRT) were measured following the procedures outlined by the American Speech-Language-Hearing Association (Schlauch et al., 1996).

2.3. Neuropsychological assessment

The Montreal Cognitive Assessment (MoCA) was administered to evaluate global cognitive function (Nasreddine et al., 2005). Domain-specific abilities were further evaluated using the Chinese version of the Auditory Verbal Learning Test (AVLT) (Zhao et al., 2012) for verbal memory, the Stroop Color and Word Test (Stroop) (Periáñez et al., 2021) for working memory, the Symbol Digit Modalities Test (SDMT) (Harand et al., 2018) for attention, and the Trail Making Test parts A and B (TMT-A, TMT-B) (Wang et al., 2021) for processing speed and executive function, respectively. The validated Hospital Anxiety and Depression Scale (HADS) was administered to assess participants’ emotional status (Zigmond and Snaith, 1983). To minimize the potential impact of hearing impairment on cognitive testing, all assessments were conducted in a quiet environment by experienced examiners. Instructions were delivered slowly and clearly, with repetition and visual clarification provided when necessary to ensure adequate task comprehension.

2.4. MRI acquisition

MRI data acquisition employed a 3Tesla Philips Achieva system with an 8-channel head coil. The specific imaging parameters for the 3D pCASL sequence were implemented in full accordance with the international consensus guidelines for aging populations (Alsop et al., 2015): repetition time (TR) = 4034.6 ms, echo time (TE) = 10.7 ms, imaging flip angle = 90°, label duration = 1800 ms, and post-labeling delay = 1800 ms. The geometric parameters included a FOV = 240 × 240 mm2, an acquisition matrix = 80 × 80, an acquired voxel size of 3.0 × 3.0 × 6.0 mm3 and 14 slices. Background suppression was applied to minimize static tissue signals and maximize perfusion contrast. The acquisition consisted of 12 temporal dynamics (12 label-control pairs). For absolute CBF quantification, a proton density-weighted reference image was automatically acquired under matched spatial conditions. In bilateral auditory cortices, volumes of interest (VOI, 40 × 30 × 20 mm3, Fig. 1) were defined with their centers located on Heschl’s gyrus (Abdul-Kareem and Sluming, 2008). GABA measurements were quantified using macromolecule-suppressed MEGA-PRESS sequences (2000 ms TR; 80 ms TE; 2000 Hz bandwidth; editing pulses: 1.9 / 1.5 ppm; 320 averages). Glu signals were collected from the same VOIs using a PRESS sequence (2000 ms TR; 35 ms TE; 2000 Hz bandwidth; 64 averages) (Li et al., 2023).

Fig. 1.

Fig. 1.

Open in a new tab

MEGA-PRESS and PRESS spectra were collected from the VOIs (40 × 30 × 20 mm3) centered on the Heschl’s gyrus in the left auditory region (A) and right auditory region (B), respectively. The black curve represents the raw data, the red curve represents the fitting data, and the shadow area represents the mean data ±1 standard deviation. MEGA-PRESS, Mescher-Garwood point-resolved spectroscopy; PRESS, Point-resolved Spectroscopy; VOI, volumes of interest.

2.5. MRI data processing

The data were processed using ExploreASL (https://zenodo.org/records/3986669), in combined with SPM12 (https://www.fil.ion.ucl.ac.uk/spm/) and CAT12 (https://neuro-jena.github.io/cat//). The processing pipeline consisted of the following steps: (1) Segment the T1-weighted structural scan into gray matter (GM), white matter (WM), and cerebrospinal fluid (CSF); Perform motion correction, outlier rejection, and registration of PCASL images to the structural space; (2) Quantify CBF based on the kinetic modeling approach outlined in the 2015 guidelines from the ISMRM perfusion study group; (3) Apply partial volume correction to the CBF maps using the T1 segmentation results; (4) Create analysis masks to exclude non-brain and artifact-laden voxels, and then perform spatial normalization to the MNI space; and (5) Compute the mean quantitative CBF values at the global GM level and in GM areas of interest defined by the MNI atlas (Easson et al., 2023).

2.6. MRS data analysis

MEGA-PRESS spectra with macromolecule suppression were processed using Gannet 3.1 (Edden et al., 2014). An exponential line broadening of 3 Hz was applied before spectral fitting, and the GABA peak near 3 ppm was modeled using a Gaussian model. PRESS spectra were analyzed using LCModel v6.3–1M to estimate Glu concentrations (Provencher, 2005). In accordance with established quality control standards, only spectra with GABA fitting errors <20% and Glu Cramér-Rao lower bounds (CRLB) <20% were included in subsequent analyses. Metabolite concentrations were adjusted for T1 and T2 relaxation times and partial volume contributions, and reported in institutional units (i. u.) based on the standardized computation formulas (Gasparovic et al., 2006; Mullins et al., 2014). To ensure rigorous quality control of the MEGA-PRESS spectral data, additional baseline technical metrics—including GABA spectral linewidth, Signal-to-Noise Ratio (SNR), unsuppressed water linewidth, and average frequency drift—were systematically evaluated for bilateral auditory regions. Additional details are provided in the Supplementary Material.

2.7. Spatial correlation analysis with neurotransmitter maps

To assess the neurotransmitter systems potentially implicated by the CBF, this study extracted spatial distribution information of various neurotransmitter receptors and other related molecular markers from the JuSpace toolbox (version 2.0, https://github.com/juryxy/JuSpace). These encompassed the serotonergic, dopaminergic, GABAergic, opioid, noradrenergic, cholinergic, and glutamatergic systems, as well as cerebral metabolic markers, inflammation-related indicators, and epigenetic regulators. Following normalization to MNI standard space, the CBF images were smoothed using an 8 mm isotropic Gaussian kernel. To investigate the relationship between CBF and neurochemistry, the spatial patterns of the processed CBF images were correlated with neurotransmitter distribution maps. Specifically, regional CBF values were quantified across 119 brain areas defined by the Neuromorphometrics Atlas (MICCAI 2012) within the JuSpace toolbox. These extracted mean values were then statistically compared to corresponding neurotransmitter densities using Fisher’s Z-transformed Spearman rank correlations. To validate the significance of these coefficients against a null distribution, permutation testing (10,000 iterations) was achieved by randomly shuffling group labels. To account for multiple comparisons, all p-values were adjusted using the False Discovery Rate (FDR) method, with the significance threshold set at p < 0.05.

2.8. Statistical analysis

Normality of continuous variables was assessed using the Kolmogorov-Smirnov test. Group-level statistical analyses employed independent two-sample t-tests for parametric data and Mann-Whitney U tests for non-parametric distributions, as determined by prior normality assessments. Normally distributed data are reported as mean ± standard deviation (SD), whereas non-normally distributed variables are presented as median with interquartile range (IQR). Sex distribution between groups was examined using chi-square tests. Partial correlation analyses, controlling for age and education, were conducted in the presbycusis group to assess associations among CBF, neurotransmitter concentrations, hearing thresholds, and cognitive performance. Statistical procedures were performed with IBM SPSS Statistics version 26.0 (IBM Corp., Armonk, NY, USA). A two-tailed p-value <0.05 was regarded as statistically significant. Importantly, correlation analyses were restricted to brain regions that exhibited group-significant differences in CBF.

Given the observed significant associations among GABA concentrations, CBF levels, hearing impairment and cognitive performance in presbycusis (see Results section), mediation analyses were carried out to test the following hypotheses: (1) whether GABA concentration mediate the association between auditory impairment and CBF; (2) whether CBF levels mediate the relationship between GABA and cognitive performance; (3) whether GABA and CBF act as sequential mediators connecting hearing impairment to cognitive decline. Mediation analyses were employed using the PROCESS macro in SPSS, employing Model 4 for single mediation and Model 6 for sequential mediation, with age and education statistically controlled. Indirect pathway estimates relied upon bias-corrected bootstrapping with 5000 iterations, and statistical significance denoted when the 95% confidence interval (CI) excluded zero.

3. Results

3.1. Participant characteristics

No significant group-level differences were observed between the presbycusis and NH participants regarding age, sex, educational background, or scores on anxiety and depression scales (all p > 0.05; Table 1). Nevertheless, presbycusis exhibited markedly reduced hearing abilities and cognitive performance in comparison to the NH group (all p < 0.01; Table 1).

Table 1.

Participants’ demographic and clinical data.

Characteristics NH group (n = 91) presbycusis group (n = 124) z/χ2/t-value NH vs. presbycusis df p-value NH vs. presbycusis
Age, years, median (IQR) 65.0 (62.0–67.0) 66.0 (62.0–69.0) −1.819 N/A 0.465
Education, years, median (IQR) 12.0 (9.0–14.0) 12.0 (9.0–12.0) −0.397 N/A 0.691
Sex (Male/Female) 48/43 72/52 0.602 1 0.438
Anxiety, median (IQR) 3.0 (1.0–5.0) 2.0 (1.0–5.0) −1.548 N/A 0.122
Depression, median (IQR) 3.0 (1.0–5.0) 3.0 (1.0–6.0) −1.000 N/A 0.318
PTA, dB/HL, median (IQR) 11.3 (8.8–15.0) 31.3 (27.5–40.0) −12.538 N/A <0.001 ***
SRT, dB/HL, median (IQR) 12.5 (9.5–14.5) 31.5 (25.5–42.5) −12.272 N/A <0.001 ***
MOCA, median (IQR) 27.0 (25.0–28.0) 24.0 (20.3–27.0) −4.559 N/A <0.001 ***
AVLT, mean (SD) 51.01 (12.17) 45.91 (11.11) 3.149 183.512 0.002 **
SDMT, mean (SD) 34.45 (11.75) 23.53 (10.87) 6.948 185.075 <0.001 ***
Stroop, s, median (IQR) 130.0 (117.0–152.0) 151.0 (132.0–179.8) −4.896 N/A <0.001 ***
TMT-A, s, median (IQR) 54.0 (45.0–78.0) 70.0 (54.0–90.0) −3.493 N/A <0.001 ***
TMT-B, s, median (IQR) 151.0 (118.0–206) 202.5 (162.0–278.3) −4.635 N/A <0.001 ***
Open in a new tab

NH, normal hearing; s, second; df, degree of freedom; IQR, interquartile range; SD, standard deviation; dB/HL, decibel hearing level; PTA, pure tone average; SRT, speech reception threshold; MOCA, Montreal cognitive assessment; AVLT, Auditory verbal learning test; SDMT, Symbol digit modalities test; TMT, Trail-making test. Values in bold are used to indicate statistical significance.

**

p < 0.01,

***

p < 0.001.

3.2. Group differences in auditory region MRS data

The quantification of GABA and Glu in both auditory cortices showed comparable fitting errors and CRLBs between the two groups, with no statistically significant differences observed (all p > 0.05; Table 2). One presbycusis was excluded due to poor spectral quality. Substantial reductions in GABA and Glu concentrations were exhibited in bilateral auditory cortices of presbycusis relative to NH participants.

Table 2.

MRS data of the NH group and presbycusis group.

Characteristics NH group (n = 91) presbycusis group (n = 124) z/t-value NH vs. presbycusis p-value NH vs. presbycusis
 Left auditory region
GABA, i.u., median (IQR) 1.0 (1.0–1.3) 0.93 (0.8–1.2) −3.959 <0.001***
GABA fitting errors, median (IQR) 9.8 (7.9–12.0) 9.7 (8.1–11.9) −0.066 0.947
Glu, i.u., median (IQR) 4.6 (4.3–5.0) 4.3 (3.9–4.8) −3.460 0.001**
Glu CRLB, median (IQR) 4.6 (4.0–5.7) 4.9 (4.2–5.9) −0.748 0.454
 Right auditory region
GABA, i.u., median (IQR) 1.3 (1.1–1.5) 1.1 (1.0–1.3) −4.224 <0.001***
GABA fitting errors, median (IQR) 9.8 (7.2–11.3) 9.0 (7.4–11.3) −0.189 0.850
Glu, i.u., median (IQR) 4.5 (4.1–4.9) 4.0 (3.6–4.6) −4.167 <0.001***
Glu CRLB, median (IQR) 5.4 (4.2–7.1) 5.6 (4.7–7.3) −1.114 0.265
Open in a new tab

MRS, magnetic resonance spectroscopy; NH, normal hearing; IQR, interquartile range; i.u., institutional units; GABA, γ-aminobutyric acid; Glu, glutamate; CRLB, Cramer-Rao lower bound. Values in bold are used to indicate statistical significance.

**

p < 0.01,

***

p < 0.001.

3.3. Brain regions with significant CBF differences

Voxel-wise analysis indicated clusters with significantly altered CBF levels between the presbycusis and NH groups, applying FDR correction (voxel-level p < 0.001; cluster-level p < 0.05). Clusters consisting of more than 10 voxels were mapped to anatomical regions using the AAL atlas. As presented in Fig. 2 and Table 3, presbycusis patients showed reduced CBF in multiple regions.

Fig. 2.

Fig. 2.

Open in a new tab

Group differences in CBF between presbycusis patients and NH controls. Regions with significantly reduced CBF in the presbycusis group are shown in warm colors. Statistical comparisons were conducted using a two-sample t-test, with results corrected for multiple comparisons using FDR correction (p < 0.001, cluster size >10 voxels). NH, normal hearing; CBF, cerebral blood flow; FDR, false discovery rate; L, left; R, right.

Table 3.

Brain regions with significant differences in the CBF between the NH group and presbycusis group.

Brain region Brodmann area MNI coordinates T value Cluster size
x y z
R Precentral 8 51 9 42 4 13
L Superior frontal gyrus 10 −15 69 9 4.51 32
R Superior frontal gyrus 10 24 57 27 4.23 94
R Middle frontal gyrus 8 48 12 45 4.26 23
L Inferior frontal gyrus, opercular part 44 −54 12 12 3.87 25
R Inferior frontal gyrus, opercular part 44 51 15 18 3.74 115
L Inferior frontal gyrus, triangular part 45 −57 21 18 4.57 71
R Inferior frontal gyrus, triangular part 45 54 24 9 4.16 68
L Inferior frontal gyrus, orbital part 38 −45 18 −12 5.02 46
R Inferior frontal gyrus, orbital part 47 36 30 −6 3.72 37
L Rolandic operculum 13 −42 −6 3 3.7 15
R Rolandic operculum 6 48 −6 6 5.52 106
L Supplementary motor area 32 −3 6 45 3.79 12
L Superior frontal gyrus, medial 9 −3 57 27 4.05 191
R Superior frontal gyrus, medial 9 3 57 27 3.90 164
L Superior frontal gyrus, medial orbital 11 −3 51 −12 4.49 63
R Superior frontal gyrus, medial orbital 11 33 48 −9 3.89 51
L Rectus 11 −3 51 −15 3.74 12
L Insula 38 −42 9 −12 4.16 135
R Insula 38 42 9 −12 5.46 305
L Anterior cingulate and paracingulate gyri 10 −3 27 27 4.75 148
R Anterior cingulate and paracingulate gyri 9 9 33 27 3.7 135
L Median cingulate and paracingulate gyri 32 −6 18 33 3.9 50
R Median cingulate and paracingulate gyri 32 3 18 36 5.13 81
R Calcarine 17 18 −87 0 4.69 37
L Cuneus 7 −6 −78 36 3.68 12
L Superior occipital gyrus 7 −24 −78 42 3.83 16
R Superior occipital gyrus 18 18 −87 18 4.38 33
L Middle occipital gyrus 19 −24 −78 36 3.97 32
R Middle occipital gyrus 39 33 −63 33 4.97 25
L Superior parietal gyrus 7 −24 −72 51 4.51 119
R Superior parietal gyrussx 7 15 −72 57 5.09 60
R Angular gyrus 40 48 −63 45 3.82 40
L Precuneus 7 −6 −78 48 4.14 120
R Precuneus 7 6 −78 48 3.94 111
L Caudate nucleus 9 −9 12 12 3.9 59
R Caudate nucleus 32 9 12 12 6.15 63
L Thalamus 41 −3 −21 3 4.26 19
R Thalamus 41 3 −21 3 4.75 19
R Heschl gyrus 22 48 −15 6 4.33 38
L Superior Temporal gyrus 13 −39 −21 3 4.4 33
R Superior Temporal gyrus 42 69 −18 9 4.83 59
L Temporal Pole: Superior Temporal gyrus 22 −57 6 0 3.79 35
R Temporal Pole: Superior Temporal gyrus 38 54 18 −9 4.58 62
Open in a new tab

FDR corrected p < 0.001, cluster size >10 voxels. AAL, anatomical automatic labeling; MNI, Montreal Neurological Institute; FDR, False Discovery Rate correction; L, left; R, right.

3.4. Associations between neurotransmitter levels and clinical scores

As presented in Fig. 3 and Supplementary Table S1, within the presbycusis group, right auditory cortex GABA levels in the presbycusis group showed positive correlations with MoCA and AVLT scores, while being negatively related to TMT-A, TMT-B, PTA, and SRT. Additionally, right auditory Glu levels showed a significant inverse relationship with PTA, while a positive association was observed between left auditory Glu concentrations and MoCA.

Fig. 3.

Fig. 3.

Open in a new tab

The relationship between neurotransmitter, CBF and clinical scores. CBF, cerebral blood flow; L, left; R, right; IFGoperc, Inferior frontal gyrus, opercular part; IFGtriang, Inferior frontal gyrus, triangular part; ORBinf, Inferior frontal gyrus, orbital part; ROL, Rolandic operculum; SFGmed, Superior frontal gyrus, medial; ORBsupmed, Superior frontal gyrus, medial orbital; INS, Insula; ACG, Anterior cingulate and paracingulate gyri; CUN, Cuneus; SOG, Superior occipital gyrus; MOG, Middle occipital gyrus; SPG, Superior parietal gyrus; ANG, Angular gyrus; PCUN, Precuneus; CAU, Caudate nucleus; THA, Thalamus; HES, Heschl’s gyrus; TPOsup, Temporal pole: superior temporal gyrus; SFG, Superior frontal gyrus; STG, Superior temporal gyrus.

3.5. Associations between CBF levels and clinical scores

Fig. 3 and Supplementary Table S2 illustrate the relationship between CBF levels and clinical measures in the presbycusis group. Lower MoCA scores were significantly related to higher CBF in several regions: bilateral superior frontal gyrus, left opercular and triangular parts of the inferior frontal gyrus, left Rolandic operculum, left thalamus, and left superior temporal gyrus. A significant positive association existed between right superior parietal CBF and TMT-A. Conversely, CBF levels in right Heschl’s gyrus were inversely associated with SRT.

3.6. Associations between neurotransmitters and CBF levels

Fig. 3 and Supplementary Table S3 present correlations between neurotransmitters and CBF levels. Presbycusis patients exhibited significant positive correlations between left auditory cortex GABA concentrations and CBF in the following regions: right opercular part of the inferior frontal gyrus, right Rolandic operculum, bilateral anterior cingulate and paracingulate gyri, bilateral caudate nucleus, right thalamus, right Heschl’s gyrus, and right temporal pole of the superior temporal gyrus.

A positive relationship was observed between right auditory cortex GABA and CBF in the following regions: right opercular, bilateral triangular and orbital parts of the inferior frontal gyrus, right Rolandic operculum, bilateral media and left orbital medial superior frontal gyrus, right insula, bilateral anterior cingulate and paracingulate gyri left cuneus, right superior and middle occipital gyrus, right superior parietal gyrus, right angular gyrus, left precuneus, bilateral caudate nucleus, bilateral thalamus, and right Heschl’s gyrus.

3.7. Mediation effects of neurotransmitters and CBF levels between auditory function and cognitive function

To test the hypothesis of a systemic neurovascular-metabolic cascade, mediation and serial mediation analyses were conducted (Figs. 4 and 5).

Fig. 4.

Fig. 4.

Open in a new tab

Simple mediation models of the associations among hearing loss, GABA levels, CBF levels, and cognitive impairment in presbycusis. (A)Mediation effects of the right auditory GABA levels on the association between the PTA and TMT-A. (B)Mediation effects of the SPG.R_CBF on the association between the right auditory GABA levels and TMT-A scores. (C)Mediation effects of the Tha. L_CBF on the association between the right auditory GABA levels and MOCA scores. L, left; R, right; GABA, γ-aminobutyric acid; CBF, cerebral blood flow; PTA, pure tone average; MOCA, Montreal cognitive assessment; TMT, Trail-making test; SPG, Superior parietal gyrus; Tha, Thalamus.

Fig. 5.

Fig. 5.

Open in a new tab

Serial mediation models of the associations among hearing loss, GABA levels, CBF levels, and cognitive impairment in presbycusis. (A)A serial mediation model including the right auditory GABA levels and SPG.R_CBF as mediators of the associations between PTA and TMT-A. (B)A serial mediation model including the right auditory GABA levels and Tha.L_CBF as mediators of the associations between PTA and TMT-A. L, left; R, right; GABA, γ-aminobutyric acid; CBF, cerebral blood flow; PTA, pure tone average; MOCA, Montreal cognitive assessment; TMT, Trail-making test; SPG, Superior parietal gyrus; Tha, Thalamus.

3.7.1. Simple mediation

Fig. 4 shows simple mediation models of the relationship between hearing impairment, GABA concentrations, CBF concentrations, and cognitive performance in patients with presbycusis. Simple mediation model A showed PTA scores significantly predicted TMT-A scores (Fig. 4A, c = 0.653, p = 0.016). Once GABA concentrations were considered, the direct link between PTA and TMT-A was rendered non-significant (c’ = 0.322, p = 0.243). Notably, the indirect pathway through GABA levels remained significant (0.331, 95% CI [0.055, 0.664]).

In Model B, a significant total effect was observed (Fig. 4B, c = 40.240, p < 0.001), indicating that right auditory GABA concentrations were predictive of TMT-A performance in individuals with presbycusis. GABA continued to exhibit a statistically significant effect on TMT-A scores, even when CBF was controlled for (c’ = −47.094, p < 0.001). Additionally, the mediation analysis revealed a significant indirect pathway through CBF levels (6.855, 95% CI [1.896, 15.562]).

In simple mediation model C (Fig. 4C), right auditory cortex GABA levels had a significant total effect on MoCA scores (c = 3.085, p = 0.025). Even after adjusting for thalamic CBF, the direct relationship remained significant (c’ = 4.008, p = 0.004). The indirect effects mediated by CBF in the thalamus were also significant (−0.9234, 95% CI [−2.084, −0.161]).

3.7.2. Serial mediation

Fig. 5 shows the serial mediation analysis examining the pathway from hearing loss to cognitive impairment through GABA and CBF in presbycusis. The serial mediation model A revealed that PTA scores also predicted TMT-A scores (Fig. 5A, c = 0.653, p = 0.016). Adjustment for the two mediators reduced the direct effect, which no longer reached significance (c’ = 0.250, p = 0.348). The overall mediation effect remained significant (0.402, 95% CI [0.082, 0.767]), with a specific indirect path through both GABA and CBF also reaching statistical significance (−0.071, 95% CI [−0.175, −0.011]).

Finally, the serial mediation model B demonstrated that PTA scores significantly predicted MOCA scores (Fig. 5B, c = −0.088, p = 0.014). The inclusion of GABA and thalamic CBF as mediators weakened the direct effect to the point of statistical insignificance (c’ = −0.065, p = 0.077). While the overall indirect effect did not reach significance (−0.023, 95% CI [−0.072, 0.015]), specific indirect pathway through GABA levels and thalamic CBF showed a significant mediating effect (0.009, 95% CI [0.001, 0.022]).

3.8. Correlation analysis between CBF and metabolic modality neurotransmitter maps

Fig. 6 and Supplementary Table S4 present the co-localization of CBF levels with the spatial distribution of specific neurotransmitter systems. In presbycusis patients as compared to NH, CBF was significantly associated with the spatial distribution of 5-hydroxytryptamine type 1b receptor, 5-hydroxytryptamine type 2a receptor, serotonin transporter, dopamine type 1 receptor, dopamine transporter, γ-aminobutyric acid type a5 receptor (GABAa5), vesicular acetylcholine transporter, cerebral metabolic rate of glucose (CMRglu), and histone deacetylase (HDAC).

Fig. 6.

Fig. 6.

Open in a new tab

Spatial correlation analysis between CBF and metabolic modality neurotransmitter maps (FDR corrected, *pFDR < 0.05). 5HT, 5-hydroxytryptamine; SERT, serotonin transporter; D1, dopamine type 1; D2, dopamine type 2; DAT, dopamine transporter; GABAa5, γ-aminobutyric acid type a5; MOR, mu opioid receptor; NAT, noradrenaline transporter; VAChT, vesicular acetylcholine transporter, mGluR5, metabotropic glutamate type 5; CB1, cannabinoid type 1; CMRglu, cerebral metabolic rate of glucose; HDAC, histone deacetylase; FDR, false discovery rate.

Furthermore, we examined the association of the above CBF-neurotransmitter co-localization strength with cognitive function. The analysis demonstrated that in the presbycusis group, CMRglu (r = −0.209, p = 0.022) and HDAC (r = −0.23, p = 0.012) were negatively associated with Stroop scores, while GABAa5 was positively associated with TMT-A scores (r = 0.181, p = 0.049).

4. Discussions

In this study, we systematically studied the interplay between auditory neurotransmitter alterations, cerebral perfusion abnormalities, and cognitive impairment in presbycusis using a multimodal neuroimaging approach. Our key findings are fivefold: (1) patients with presbycusis exhibited bilateral reductions in auditory GABA and Glu levels; (2) widespread decreases in CBF were observed across auditory, salience, cognitive control, and visual networks; (3) right auditory GABA and regional CBF in the right superior parietal gyrus and left thalamus were significantly associated with cognitive performance; (4) mediation analysis indicated that GABA and CBF jointly mediated the correlation between hearing impairment and cognitive deterioration; and (5) spatial correlation analyses revealed significant co-localization between CBF alterations and multiple neurotransmitter systems, further relating to cognitive scores. Collectively, these findings support a neurovascular-metabolic association framework linking auditory dysfunction to cognitive impairment in presbycusis, spanning molecular, vascular, and functional domains.

Our findings demonstrated right Heschl’s gyrus CBF levels were decreased in presbycusis and show a negative association with speech recognition performance. Heschl’s gyrus, a key hub within the auditory processing network, is essential for the perception and integration of sound. Reduced CBF levels in this region may impair auditory integration and contribute to deficits in speech comprehension (Silva et al., 2020). Our results align with previous studies, including an ASL study that demonstrated substantial perfusion in the right auditory cortex, particularly targeting Heschl’s gyrus, in patients diagnosed with sensorineural hearing loss (Ponticorvo et al., 2019). Furthermore, a study using 18F-FDG PET revealed right auditory cortex glucose metabolic activity was reduced in the among individuals with hearing impairment (Verger et al., 2017). These findings indicate that hearing impairment results in reduced auditory input, thereby decreasing neuronal activity and metabolic demand within the auditory cortex. This decline in functional engagement was corroborated by a positive correlation between auditory metabolic activity and Mini-Mental State Examination scores, underscoring the pivotal involvement of the auditory cortex in speech perception as well as broader cognitive function.

Importantly, CBF abnormalities in presbycusis were not confined to auditory regions alone. Reduced perfusion was also observed in critical regions of the salience network, notably the insula and opercular inferior frontal gyrus. This network is essential for coordinating the dynamic interplay between external sensory signals and internal attentional mechanisms. Decreased CBF levels in these regions may disrupt the brain’s capacity to identify relevant auditory or visual cues, hinder attentional shifting, and compromise contextual adaptation. According to Menon’s theoretical model, the anterior insula serves as a “network switch”, enabling transitions between internally focused self-referential cognition and externally directed goal-oriented processing (Menon and Uddin, 2010). Hypoperfusion in this region may disrupt its switching function, contributing to diminished responsiveness to novel stimuli and potentially explaining the co-occurrence of attentional deficits and cognitive deterioration in those with hearing impairment. Additionally, this study demonstrated reduced CBF levels in higher-order cognitive regions, such as the superior frontal gyrus, which was significantly associated with cognitive dysfunction. We also observed decreased perfusion in the superior occipital gyrus among presbycusis patients. Our findings indicate that hearing impairment might lead to functional decline within the auditory system and trigger compensatory mechanisms in other sensory networks, such as the visual network (Slade et al., 2020). The superior occipital gyrus, which is critical for visual information processing, may take on additional perceptual responsibilities in the context of auditory decline. This increased functional demand may eventually lead to reduced local CBF levels.

Additionally, we observed decreased concentrations of GABA and Glu in bilateral auditory cortices. The central nervous system’s primary inhibitory neurotransmitter, GABA, is critically involved in sound information processing (Caspary et al., 1995), and its reduction may reflect impaired inhibitory control within auditory pathways. In contrast, glutamate, the principal excitatory neurotransmitter, is fundamental to auditory transmission and integration. The observed decrease in Glu suggests compromised excitatory neuronal activity in the central auditory system (Li et al., 2023). Notably, in contrast to previous studies that reported unilateral changes, our findings of bilateral reductions in GABA and Glu levels indicate a more widespread effect of presbycusis (Wang et al., 2024). Furthermore, lower GABA concentrations were significantly linked to poorer performance in speech recognition, slower cognitive processing, and weakened executive control. The mediation analysis identified right auditory GABA as a critical mediator in the pathway connecting hearing impairment to cognitive dysfunction. Supporting this, previous findings have consistently shown that lower right auditory cortex GABA concentrations are strongly associated with auditory deficits in presbycusis (Gao et al., 2015; Li et al., 2023). However, an animal study in C57BL/6 mice exhibiting progressive hearing loss found increased GABA receptor expression in the auditory cortex and temporal lobe, accompanied by worsened spatial memory but enhanced cognitive flexibility (Beckmann et al., 2020). These findings reinforce the notion that auditory deficits may influence both neural architecture and cognitive performance, supporting the mediation mechanisms proposed in our model.

Further analysis revealed significant positive correlations between the right auditory cortex GABA concentration and CBF in the left thalamus and right superior parietal gyrus. This may reflect neurotransmitter-driven neurometabolic-vascular coupling mechanisms. Previous studies suggest that GABAergic interneurons regulate local excitation-inhibition balance and cerebral perfusion (Heckers and Konradi, 2015). Modinos et al. reported that decreased GABA in the medial prefrontal cortex was significantly associated with hippocampal hyperperfusion at rest in ultra-high risk psychosis, particularly among those who subsequently transitioned to psychosis (Modinos et al., 2018). This indicates that GABAergic inhibitory deficits may lead to disinhibition of glutamatergic neurons, thereby increasing local metabolic demands and triggering compensatory upregulation of blood flow (Stone et al., 2012). Based on this evidence, we speculate that in patients with presbycusis, alterations in CBF may be influenced by neurotransmitter levels, particularly through mechanisms mediated by the GABAergic system.

Simple model B showed that CBF levels in the left thalamus served as a mediator between right auditory GABA concentrations and overall cognitive performance in individuals with presbycusis. Integral to the auditory circuitry, the thalamus is responsible for the transmission and integration of auditory information. Earlier research has indicated that thalamic dysfunction in older individuals with hearing impairment may hinder efficient processing of auditory signals (Peelle et al., 2011). Furthermore, the thalamus serves as a critical node in multiple functional circuits supporting cognition, including memory, attention, and information processing (Zhu et al., 2017). Xu et al. (Xu et al., 2019) reported that sensorineural hearing loss individuals showed reduced spontaneous neural activity in the thalamus, accompanied by widespread disruptions in its functional connectivity with other brain regions. These changes showed a positive correlation with impairments in verbal learning and memory, suggesting a contributory role of thalamic dysfunction in cognitive deterioration. Consistent with these findings, our study observed reduced CBF in the thalamus, which was inversely correlated with global cognitive performance. This association suggests that altered thalamic perfusion may be linked to impaired neuronal activity and cognitive dysfunction in presbycusis. Mediation model C demonstrated that the right superior parietal gyrus CBF levels acted as a mediator between right auditory GABA concentrations and cognitive processing speed. These findings support a potential association between regional hemodynamic alterations, neurochemical changes, and cognitive dysfunction. The superior parietal gyrus, a crucial region within the parietal lobe, contributes to advanced cognitive functions including attention, multisensory integration, and spatial cognition (Stoeckel et al., 2004). In our study, decreased CBF levels in the right superior parietal gyrus were linked to slower information processing in individuals with presbycusis. Consistent with this, previous studies on mild cognitive impairment have reported decreased CBF levels in the left superior parietal gyrus, which were negatively related to Mini-Mental State Examination scores (Qiu et al., 2023). These results highlight the potential role of perfusion abnormalities in this region as a key contributor to cognitive decline. Importantly, serial mediation analyses revealed that altered GABA concentrations and CBF levels were statistically associated with the relationship between hearing impairment and cognitive decline. However, given the cross-sectional nature of the study, these findings should be interpreted as association patterns rather than evidence of definitive causal pathways. This finding emphasizes that hearing impairment in presbycusis might not only directly affect cognition through imbalances in auditory neurotransmitter levels but also have indirect effects via reduced cerebrovascular perfusion in auditory pathways and associated regions.

Our analysis demonstrated that CBF alterations in presbycusis robustly co-localized with several neurotransmitter systems, implicating monoaminergic, cholinergic, GABAergic, and metabolic pathways. This overlap, particularly with serotonin and dopamine circuits, suggests that hearing loss disrupts neuromodulator systems essential for higher-order cognitive functions commonly impaired in this condition. Critically, we also found that these neurotransmitter maps were directly correlated with cognitive performance. Negative correlations of CMRglu and HDAC with Stroop scores point to metabolic and epigenetic contributions to inhibitory control deficits. Meanwhile, the positive correlation between GABAa5 and TMT-A performance highlights the importance of GABAergic signaling in maintaining processing speed. These results complement our mediation models by revealing putative molecular mechanisms that bridge neurovascular dysfunction and cognition, emphasizing the value of a multimodal approach to decipher the pathway from presbycusis to cognitive impairment.

The findings of this study converge into a systemic neurovascular-metabolic cascade (Fig. 7). Specifically, the chronic deprivation of auditory input triggers a reduction in inhibitory and excitatory neurotransmission within the auditory cortex. This metabolic downregulation, via neurovascular coupling, subsequently leads to extensive hypoperfusion in both auditory and higher-order cognitive networks. Our spatial correlation analysis further suggests that these perfusion deficits are molecularly anchored to the distribution of GABAa5 and CMRglu, providing a potential link between regional vascular failure and systematic cognitive impairment in presbycusis.

Fig. 7.

Fig. 7.

Open in a new tab

Schematic of the neurovascular-metabolic cascade linking presbycusis to cognitive impairment. GABA, γ-aminobutyric acid; CBF, Cerebral Blood Flow; HDAC, Histone Deacetylase; GABAa5, γ-aminobutyric acid type a5; CMRglu, cerebral metabolic rate of glucose.

This study is subject to several notable limitations. First, the cross-sectional design limits the ability to establish causal or temporal relationships among hearing loss, neurotransmitter alterations, cerebral perfusion abnormalities, and cognitive decline. Therefore, the proposed neurovascular-metabolic cascade should be interpreted as an association framework rather than definitive evidence of causal directionality. In addition, we did not include structural vascular markers (e.g., white matter hyperintensities) or neurodegenerative biomarkers such as amyloid/tau status and hippocampal atrophy measures. Thus, alternative explanations involving shared cerebrovascular or neurodegenerative mechanisms cannot be excluded, and the observed neurotransmitter and perfusion abnormalities may partly reflect downstream effects of common pathological aging processes rather than changes directly initiated by hearing loss. Second, although adaptive procedures were implemented during neuropsychological assessments, including repeated instructions and visual clarification when necessary, we cannot completely exclude the possibility that reduced auditory access may have partially influenced performance on verbally mediated cognitive tasks. Third, the relatively limited representation of individuals with severe hearing loss may reduce statistical power and limit the generalizability of the findings. Fourth, the relatively large auditory VOIs used in the MRS analysis may have included adjacent association cortex and white matter, potentially introducing partial-volume effects and limiting spatial specificity. Although tissue-fraction correction was performed, the absence of a non-auditory control VOI remains a methodological limitation. Furthermore, macromolecule-suppressed GABA editing is sensitive to field instability and frequency drift, although extensive quality-control procedures were applied to minimize these effects. Finally, ASL perfusion was quantified using a single post-labeling delay, and potential group- or region-specific differences in arterial transit time may have influenced the measured ASL signal independently of true cerebral blood flow. In addition, physiological and vascular factors that may affect ASL-derived CBF measurements, including blood pressure, CO2 levels, hematocrit, and vasoactive medication use, were not systematically assessed or controlled for. Future longitudinal studies incorporating multimodal vascular and neurodegenerative biomarkers, multi-delay ASL protocols, and larger cohorts are needed to further validate and extend the present findings.

5. Conclusions

This research delved into the metabolic correlation linking hearing loss to cognitive decline in presbycusis. The results indicated that changes in the right auditory cortex GABA concentrations mediated the relationship between auditory deficits and cognitive decline. Moreover, lowered CBF levels in the right superior parietal gyrus and left thalamus served as mediators between right auditory GABA concentrations and cognitive performance. More significantly, the combined effects of altered right auditory cortex GABA concentration and reduced CBF levels in the right superior parietal gyrus or left thalamus jointly mediated the correlation between hearing impairment and cognitive deterioration. Further spatial correlation analysis revealed that CBF alterations in presbycusis significantly co-localization with multiple neurotransmitter systems, cerebral glucose metabolism and epigenetic markers. Crucially, the identified correlations between these neurotransmitter maps and cognitive performance provide additional validation of the functional relevance of these spatial couplings. These findings not only reinforce the pivotal mediating roles of GABA and cerebral blood flow in the hearing-cognition relationship but also contextualize these changes within a broader neurochemical framework. This research identifies a unique neurometabolic mechanism connecting hearing impairment with cognitive deficits, highlighting potential biomarkers for cognitive impairment in presbycusis. Therapeutic strategies aimed at rebalancing inhibitory neurotransmission within the auditory cortex and enhancing regional cerebral perfusion may provide novel targets for treating cognitive decline associated with presbycusis.

Supplementary Material

1

Acknowledgements

This work was supported by the National Natural Science Foundation of China (Nos. 61701342, 81601479); Taishan Scholars Project of Shandong Province (No. tstp20240525); Postdoctoral Innovation Projects of Shandong Province (Nos. SDCXZG-202303058, SDCX-ZG-202400047); Shandong Provincial Natural Science Foundation of China (Nos. ZR2025MS1359, ZR2025MS1482); Natural Science Foundation of Tianjin Municipality (No. 19JCQNJC13100); and Tianjin Natural Science Foundation Joint Fund (No. 25JCLMJC00360). This work was also supported by National Institutes of Health (Grants. R01 EB016089, R01 EB023963, R21 AG060245, P41 EB015909, and P41 EB031771).

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi.org/10.1016/j.nbd.2026.107535.

Footnotes

CRediT authorship contribution statement

Yao Wang: Conceptualization, Funding acquisition, Methodology, Project administration, Writing - review and editing. Shuya Wang: Formal analysis, Investigation, Visualization, and Writing - original draft. Wen Ma: Formal analysis, Investigation, Visualization, and Writing - original draft. Wenqing Li: Formal analysis, Investigation. Weilong Fu: Software, Validation. Fuxin Ren: Data curation, Funding acquisition, Software. Bing Liu: Software, Supervision, Writing - review and editing. Richard A.E. Edden: Resources, Software, Supervision. Muwei Li: Methodology, Software, Supervision. Xinjuan Zhang: Conceptualization, Methodology, Project administration, Writing - review and editing. Fei Gao: Conceptualization, Funding acquisition, Methodology, Project administration, Resources, Supervision, Writing - review and editing.

Ethics approval statement

Approval for this research was granted by the Institutional Review Board of Shandong University (approval number: 2016-KY-059), with all subjects signing informed consent forms before participation.

Declaration of competing interest

The authors report no competing interests.

Data availability

The data that support the findings of this study are available on request from the corresponding author. The data is not publicly available due to privacy or ethical restrictions.

References

  1. Abdul-Kareem IA, Sluming V, 2008. Heschl gyrus and its included primary auditory cortex: structural MRI studies in healthy and diseased subjects. J. Magn. Reson. Imaging 28, 287–299. 10.1002/jmri.21445. [DOI] [PubMed] [Google Scholar]
  2. Alsop DC, et al. , 2015. Recommended implementation of arterial spin-labeled perfusion MRI for clinical applications: a consensus of the ISMRM perfusion study group and the European consortium for ASL in dementia. Magn. Reson. Med. 73, 102–116. 10.1002/mrm.25197. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Bartos M, et al. , 2007. Synaptic mechanisms of synchronized gamma oscillations in inhibitory interneuron networks. Nat. Rev. Neurosci. 8, 45–56. 10.1038/nrn2044. [DOI] [PubMed] [Google Scholar]
  4. Beckmann D, et al. , 2020. Hippocampal synaptic plasticity, spatial memory, and neurotransmitter receptor expression are profoundly altered by gradual loss of hearing ability. Cereb. Cortex 30, 4581–4596. 10.1093/cercor/bhaa061. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Caspary DM, et al. , 1995. Central auditory aging: GABA changes in the inferior colliculus. Exp. Gerontol. 30, 349–360. 10.1016/0531-5565(94)00052-5. [DOI] [PubMed] [Google Scholar]
  6. Chen YC, et al. , 2020. Directed functional connectivity of the hippocampus in patients with presbycusis. Brain Imaging Behav. 14, 917–926. 10.1007/s11682-019-00162-z. [DOI] [PubMed] [Google Scholar]
  7. Chen Y, et al. , 2022. Cerebral blood flow pattern changes in unilateral sudden sensorineural hearing loss. Front. Neurosci. 16, 856710. 10.3389/fnins.2022.856710. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. De Vis JB, et al. , 2018. Arterial-spin-labeling (ASL) perfusion MRI predicts cognitive function in elderly individuals: a 4-year longitudinal study. J. Magn. Reson. Imaging 48, 449–458. 10.1002/jmri.25938. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Delano PH, et al. , 2020. Reduced suprathreshold auditory nerve responses are associated with slower processing speed and thinner temporal and parietal cortex in presbycusis. PLoS One 15, e0233224. 10.1371/journal.pone.0233224. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Donahue MJ, et al. , 2010. Baseline GABA concentration and fMRI response. Neuroimage 53, 392–398. 10.1016/j.neuroimage.2010.07.017. [DOI] [PubMed] [Google Scholar]
  11. Dukart J, et al. , 2021. JuSpace: a tool for spatial correlation analyses of magnetic resonance imaging data with nuclear imaging derived neurotransmitter maps. Hum. Brain Mapp. 42, 555–566. 10.1002/hbm.25244. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Easson K, et al. , 2023. Sex-specific cerebral blood flow alterations in youth operated for congenital heart disease. J. Am. Heart Assoc. 12, e028378. 10.1161/JAHA.122.028378. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Eckert MA, et al. , 2012. Auditory cortex signs of age-related hearing loss. J. Assoc. Res. Otolaryngol. 13, 703–713. 10.1007/s10162-012-0332-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Edden RA, et al. , 2014. Gannet: a batch-processing tool for the quantitative analysis of gamma-aminobutyric acid-edited MR spectroscopy spectra. J. Magn. Reson. Imaging 40, 1445–1452. 10.1002/jmri.24478. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Gao F, et al. , 2015. Decreased auditory GABA+ concentrations in presbycusis demonstrated by edited magnetic resonance spectroscopy. Neuroimage 106, 311–316. 10.1016/j.neuroimage.2014.11.023. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Gasparovic C, et al. , 2006. Use of tissue water as a concentration reference for proton spectroscopic imaging. Magn. Reson. Med. 55, 1219–1226. 10.1002/mrm.20901. [DOI] [PubMed] [Google Scholar]
  17. Guan B, et al. , 2022. Reorganized brain functional network topology in presbycusis. Front. Aging Neurosci. 14, 905487. 10.3389/fnagi.2022.905487. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Harand C, et al. , 2018. Evidence of attentional impairments using virtual driving simulation in multiple sclerosis. Mult. Scler. Relat. Disord. 25, 251–257. 10.1016/j.msard.2018.08.005. [DOI] [PubMed] [Google Scholar]
  19. Heckers S, Konradi C, 2015. GABAergic mechanisms of hippocampal hyperactivity in schizophrenia. Schizophr. Res. 167, 4–11. 10.1016/j.schres.2014.09.041. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Huang Q, Tang J, 2010. Age-related hearing loss or presbycusis. Eur. Arch. Otorrinolaringol. 267, 1179–1191. 10.1007/s00405-010-1270-7. [DOI] [PubMed] [Google Scholar]
  21. Lalwani P, et al. , 2019. Neural distinctiveness declines with age in auditory cortex and is associated with auditory GABA levels. Neuroimage 201, 116033. 10.1016/j.neuroimage.2019.116033. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Li N, et al. , 2023. Neurochemical and functional reorganization of the cognitive-ear link underlies cognitive impairment in presbycusis. Neuroimage 268, 119861. 10.1016/j.neuroimage.2023.119861. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Lin FR, Albert M, 2014. Hearing loss and dementia - who is listening? Aging Ment. Health 18, 671–673. 10.1080/13607863.2014.915924. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Lin FR, et al. , 2011a. Hearing loss and cognition in the Baltimore longitudinal study of aging. Neuropsychology 25, 763–770. 10.1037/a0024238. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Lin FR, et al. , 2011b. Hearing loss and incident dementia. Arch. Neurol. 68, 214–220. 10.1001/archneurol.2010.362. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Lin FR, et al. , 2011c. Hearing loss prevalence and risk factors among older adults in the United States. J. Gerontol. A Biol. Sci. Med. Sci. 66, 582–590. 10.1093/gerona/glr002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Livingston G, et al. , 2020. Dementia prevention, intervention, and care: 2020 report of the lancet commission. Lancet 396, 413–446. 10.1016/S0140-6736(20)30367-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Loughrey DG, et al. , 2018. Association of age-related hearing loss with cognitive function, cognitive impairment, and dementia: a systematic review and meta-analysis. JAMA Otolaryngol. Head Neck Surg. 144, 115–126. 10.1001/jamaoto.2017.2513. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Menon V, Uddin LQ, 2010. Saliency, switching, attention and control: a network model of insula function. Brain Struct. Funct. 214, 655–667. 10.1007/s00429-010-0262-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Modinos G, et al. , 2018. Prefrontal GABA levels, hippocampal resting perfusion and the risk of psychosis. Neuropsychopharmacology 43, 2652–2659. 10.1038/s41386-017-0004-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Mullins PG, et al. , 2014. Current practice in the use of MEGA-PRESS spectroscopy for the detection of GABA. NeuroImage 86, 43–52. 10.1016/j.neuroimage.2012.12.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Nasreddine ZS, et al. , 2005. The Montreal cognitive assessment, MoCA: a brief screening tool for mild cognitive impairment. J. Am. Geriatr. Soc. 53, 695–699. 10.1111/j.1532-5415.2005.53221.x. [DOI] [PubMed] [Google Scholar]
  33. Paciello F, et al. , 2026. Beyond the auditory system: cognitive implications of age-related hearing loss. Front. Aging Neurosci. 17. 10.3389/fnagi.2025.1736579. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Peelle JE, et al. , 2011. Hearing loss in older adults affects neural systems supporting speech comprehension. J. Neurosci. 31, 12638–12643. 10.1523/jneurosci.2559-11.2011. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Periáñez JA, et al. , 2021. Construct validity of the Stroop color-word test: influence of speed of visual search, verbal fluency, working memory, cognitive flexibility, and conflict monitoring. Arch. Clin. Neuropsychol. 36, 99–111. 10.1093/arclin/acaa034. [DOI] [PubMed] [Google Scholar]
  36. Ponticorvo S, et al. , 2019. Cortical pattern of reduced perfusion in hearing loss revealed by ASL-MRI. Hum. Brain Mapp. 40, 2475–2487. 10.1002/hbm.24538. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Profant O, et al. , 2013. Metabolic changes in the auditory cortex in presbycusis demonstrated by MR spectroscopy. Exp. Gerontol. 48, 795–800. 10.1016/j.exger.2013.04.012. [DOI] [PubMed] [Google Scholar]
  38. Provencher SW, 2005. Estimation of metabolite concentrations from localized in vivo proton NMR spectra. Magn. Reson. Med. 30, 672–679. 10.1002/mrm.1910300604. [DOI] [PubMed] [Google Scholar]
  39. Qiu M, et al. , 2023. Alterations of cerebral blood flow and its connectivity patterns measured with arterial spin Labeling in mild cognitive impairment. Curr. Alzheimer Res. 20, 567–576. 10.2174/0115672050241163231017073139. [DOI] [PubMed] [Google Scholar]
  40. Qiu X, et al. , 2024. Association between hearing ability and cortical morphology in the elderly: multiparametric mapping, cognitive relevance, and neurobiological underpinnings. eBioMedicine 104. 10.1016/j.ebiom.2024.105160. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Ren F, et al. , 2018. Gray matter atrophy is associated with cognitive impairment in patients with presbycusis: a comprehensive morphometric study. Front. Neurosci. 12. 10.3389/fnins.2018.00744. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Ren F, et al. , 2021. Brain frequency-specific changes in the spontaneous neural activity are associated with cognitive impairment in patients with presbycusis. Front. Aging Neurosci. 13. 10.3389/fnagi.2021.649874. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Schlauch RS, et al. , 1996. Identification of Pseudohypacusis using speech recognition thresholds. Ear Hear. 17, 229–236. 10.1097/00003446-199606000-00006. [DOI] [PubMed] [Google Scholar]
  44. Silva J, et al. , 2020. Cortical deafness of following bilateral temporal lobe stroke. J. Stroke Cerebrovasc. Dis. 29. 10.1016/j.jstrokecerebrovasdis.2020.104827. [DOI] [PubMed] [Google Scholar]
  45. Slade K, et al. , 2020. The effects of age-related hearing loss on the brain and cognitive function. Trends Neurosci. 43, 810–821. 10.1016/j.tins.2020.07.005. [DOI] [PubMed] [Google Scholar]
  46. Stoeckel MC, et al. , 2004. Left and right superior parietal lobule in tactile object discrimination. Eur. J. Neurosci. 19, 1067–1072. 10.1111/j.0953-816X.2004.03185.x. [DOI] [PubMed] [Google Scholar]
  47. Stone JM, et al. , 2012. Ketamine effects on brain GABA and glutamate levels with 1H-MRS: relationship to ketamine-induced psychopathology. Mol. Psychiatry 17, 664–665. 10.1038/mp.2011.171. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Su M, et al. , 2024. Alterations of excitation–inhibition balance and brain network dynamics support sensory deprivation theory in presbycusis. Hum. Brain Mapp. 45. 10.1002/hbm.70067. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Sumner P, et al. , 2010. More GABA, less distraction: a neurochemical predictor of motor decision speed. Nat. Neurosci. 13, 825–827. 10.1038/nn.2559. [DOI] [PubMed] [Google Scholar]
  50. van Dinther M, et al. , 2023. Lower cerebral blood flow predicts cognitive decline in patients with vascular cognitive impairment. Alzheimers Dement. 20, 136–144. 10.1002/alz.13408. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Verger A, et al. , 2017. Changes of metabolism and functional connectivity in late-onset deafness: evidence from cerebral 18F-FDG-PET. Hear. Res. 353, 8–16. 10.1016/j.heares.2017.07.011. [DOI] [PubMed] [Google Scholar]
  52. Wang K, et al. , 2021. Genome-wide association study identified INSC gene associated with trail making test part a and Alzheimer’s disease related cognitive phenotypes. Prog. Neuro-Psychopharmacol. Biol. Psychiatry 111. 10.1016/j.pnpbp.2021.110393. [DOI] [PubMed] [Google Scholar]
  53. Wang Y, et al. , 2024. Aberrant auditory metabolite levels and topological properties are associated with cognitive decline in presbycusis patients. Cereb. Cortex 34. 10.1093/cercor/bhae181. [DOI] [PubMed] [Google Scholar]
  54. Xu X, et al. , 2019. Sensorineural hearing loss and cognitive impairments: contributions of thalamus using multiparametric MRI. J. Magn. Reson. Imaging 50, 787–797. 10.1002/jmri.26665. [DOI] [PubMed] [Google Scholar]
  55. Xu Y, et al. , 2024. Topological disruption of low- and high-order functional networks in presbycusis. Brain Commun. 6. 10.1093/braincomms/fcae119. [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Zhao Q, et al. , 2012. Short-term delayed recall of auditory verbal learning test is equivalent to long-term delayed recall for identifying amnestic mild cognitive impairment. PLoS One 7. 10.1371/journal.pone.0051157. [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Zhu J, et al. , 2017. Altered coupling between resting-state cerebral blood flow and functional connectivity in schizophrenia. Schizophr. Bull. 43, 1363–1374. 10.1093/schbul/sbx051. [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Zigmond AS, Snaith RP, 1983. The hospital anxiety and depression scale. Acta Psychiatr. Scand. 67, 361–370. 10.1111/j.1600-0447.1983.tb09716.x. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS2208318-supplement-1.docx (44.6KB, docx)

Data Availability Statement

The data that support the findings of this study are available on request from the corresponding author. The data is not publicly available due to privacy or ethical restrictions.

RESOURCES