Altered auditory attention and functional connectivity in the auditory cortex of children with Attention-Deficit/Hyperactivity Disorder

Abstract

Background and aims

Previous studies have identified both auditory attention deficits and functional connectivity abnormalities in auditory brain regions in children with ADHD. However, auditory attention deficits are prevalent in children with Attention Deficit/Hyperactivity Disorder (ADHD), but the underlying neurophysiological mechanisms remain unclear, particularly the direct relationship between auditory attention performance and auditory cortical connectivity, which our study uniquely addresses. This study aimed to investigate the relationship between auditory attention deficits and functional connectivity (FC) in the auditory cortex and other auditory-related regions in children with ADHD.

Methods

We assessed auditory attention in 42 unmedicated children with ADHD and 36 healthy controls (HC) using the Integrated Visual and Auditory Continuous Performance Test (IVA-CPT), along with clinical symptom ratings and neuropsychological assessments. We then conducted a seed-based functional connectivity analysis, focusing on the auditory cortex and related regions. We analyzed the FC of these regions for correlations with auditory attention levels and clinical symptom scores in ADHD.

Results

Compared to HC, children with ADHD performed more poorly on the IVA-CPT, showing decreased FC between the insula, right planum polare, and bilateral cerebellum; between the left Heschl’s gyrus, right insula, and left superior temporal gyrus; and between the planum temporale and the right insula. In contrast, increased FC was observed between the right Heschl’s gyrus and the superior frontal gyrus, and between the bilateral middle frontal gyrus, right supramarginal gyrus, and right planum temporale.

Conclusions

Our results indicate that children with ADHD exhibit significant auditory attention deficits, along with numerous abnormalities in functional connectivity within the auditory cortex and related regions, which are correlated with auditory attention performance. Abnormalities in auditory cortical FC in children with ADHD may underlie auditory attention deficits. These regions could serve as potential therapeutic targets for clinical interventions aimed at ADHD children with auditory attention deficits.

Clinical trial number

Not applicable.

Introduction

Attention Deficit/Hyperactivity Disorder (ADHD) is a common neurodevelopmental disorder influenced by multiple factors. Recent comprehensive reviews indicate that ADHD affects 7.6-8.0% of children and adolescents worldwide [1, 2], representing a significant global health burden that is higher than previously estimated, and symptoms often persist into adulthood in most affected individuals [3, 4]. The core symptoms of ADHD include inattention and hyperactive/impulsive behaviors, which are categorized into three subtypes: inattentive (ADHD-PI), hyperactive-impulsive (ADHD-HI), and combined (ADHD-C). Research indicates that the core symptoms of ADHD are strongly associated with deficits in executive functions, including working memory and response inhibition [5]. Children with ADHD exhibit significant challenges across multiple domains, such as emotional regulation, academic performance, and impulse control [6, 7], which impose substantial burdens on both families and society. In addition to the core symptoms, recent studies have increasingly highlighted abnormal sensory processing in individuals with ADHD; however, the underlying neurophysiological mechanisms remain poorly understood.

Sensory processing refers to the process by which the nervous system receives, integrates, and organizes responses to external and internal stimuli [8]. Studies have identified sensory processing abnormalities in individuals with ADHD, observed in both children and adult [9–11]. A recent comprehensive meta-analysis by Jurek et al. [12] analyzing 30 studies with 5,374 participants revealed that individuals with ADHD exhibit significantly more severe sensory processing atypicalities than healthy controls across all major sensory modulation domains, with large effect sizes ranging from 1.15 to 1.23. These findings confirm that sensory processing difficulties are substantially prevalent in ADHD populations, supporting earlier observations by Mimouni-Bloch et al. [13] who found atypical sensory processing in approximately half of children with ADHD. Additionally, a recent study evaluated sensory processing in adolescents with ADHD using the Adolescent Sensory Profile, confirming that sensory processing related to smell, taste, vision, touch, and auditory stimuli significantly differed from that of neurotypical individuals [14].

Abnormalities in auditory processing are particularly evident in individuals with ADHD and may persist into adulthood [11]. Auditory attention refers to the ability to selectively focus on a specific sound source or target tone while suppressing distracting stimuli during the reception and processing of auditory information [15]. Previous studies have demonstrated deficits in both auditory selective attention and sustained attention in ADHD populations. Specifically, Studies have shown that individuals with ADHD are more sensitive to sound and more susceptible to distraction by external noises [11, 16]. Compared with neurotypical controls, ADHD patients show poorer performance on auditory selective attention tasks, particularly in noisy environments, with impaired discrimination accuracy and increased vulnerability to auditory interference [17]. Furthermore, they require greater cognitive resources to complete auditory attention tasks under identical conditions [18]. Standardized assessments of sustained auditory attention, including the Integrated Visual and Auditory Continuous Performance Test (IVA-CPT) [19], Test of Variables of Attention [20], and Conners’ Continuous Auditory Test of Attention [21], consistently reveal significantly lower scores in ADHD cohorts.

As one of the core symptoms of ADHD, attention deficit exerts non-negligible impacts on cognitive processing. Accumulating evidence indicates that attention plays a pivotal role in sensory modulation, particularly demonstrating top-down regulatory effects on sensory processing during task execution or exposure to complex stimuli [22, 23]. Among these, the frontal lobe and the brain network in which it is located play a major role in attentional regulation [23]. With advancements in resting-state functional magnetic resonance imaging (rs-fMRI), ADHD has been increasingly as a disorder of brain network dysregulation. Current pathophysiological models propose that aberrant connectivity patterns—both within and between intrinsic brain networks—may constitute the neural basis of ADHD [24]. Current studies on the auditory network have primarily focused on adults [25]. Diffusion magnetic resonance imaging (MRI) has revealed increased connectivity between the primary auditory cortex (Heschl’s gyrus) and parabelt regions, along with altered frontotemporal network integrity in individuals with ADHD [26]. Additionally, resting-state functional magnetic resonance studies have identified abnormal functional connectivity (FC) between the auditory network, Default Mode Network (DMN), and the ventral attention/significance network (VA/SN) in adults with ADHD, potentially affecting auditory attentional function [27].

The auditory cortex, a key component of the auditory network, plays a vital role in auditory processing, making its study essential. The auditory network primarily comprises the primary auditory cortex localized in Heschl’s gyrus (HG), the secondary auditory cortex within the superior temporal gyrus (STG), and associated regions including the planum polare (PP), planum temporale (PT), and insula [28, 29]. Studies have shown that these core auditory regions exhibit abnormal structural and functional connectivity in children with ADHD. Structural MRI has revealed that both the left and right Heschl’s gyrus show reduced volume, while the temporal plane is enlarged in children with ADHD compared to neurotypical children [30–32]. In resting-state functional magnetic resonance studies, children with ADHD showed a significant increase in functional connectivity between the default mode network and the auditory cortex [33, 34]. Similarly, abnormal structural and functional connectivity in the primary auditory cortex was observed in two studies on adults with ADHD [25, 26].

While prior studies have separately documented auditory attention deficits and aberrant functional connectivity in auditory cortices in children with ADHD, a critical knowledge gap remains: no study has directly examined whether auditory attention performance correlates with functional connectivity abnormalities in auditory cortex regions. Our study uniquely addresses this gap by directly correlating auditory attention performance (IVA-CPT) with seed-based functional connectivity in core auditory regions, providing the first evidence linking behavioral auditory attention deficits to specific neural connectivity abnormalities in pediatric ADHD. First, we assessed auditory attention performance using the IVA-CPT, a standardized neuropsychological tool that quantifies sustained attention through combined visual-auditory modalities and aids in ADHD diagnosis severity stratification [19, 35–38]. Additionally, Seed-based functional connectivity analysis were conducted to identify auditory cortex FC abnormalities. Finally, correlational analyses were performed to examine relationships between IVA-CPT auditory attention metrics and aberrant FC patterns. Based on the reviewed literature, we formulated specific hypotheses: (1) Children with ADHD would exhibit significant auditory attention deficits on the IVA-CPT task [37, 39]. (2) We predicted decreased functional connectivity within core auditory regions (Heschl’s gyrus, superior temporal gyrus) and increased connectivity between auditory cortex and default mode network regions, based on previous structural and resting-state findings [30–34]. (3) We hypothesized that auditory attention deficits would correlate with the magnitude of these functional connectivity abnormalities.

Materials and methods

Participants

We recruited 65 first-onset, unmedicated children with ADHD from the outpatient mental health clinic at the First Affiliated Hospital of Wenzhou Medical University. ADHD diagnoses were made by two clinically experienced psychiatrists based on the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV) criteria and the Schedule for Affective Disorders and Schizophrenia for School-Aged Children (K-SADS-PL) [40]. Exclusion criteria were: (1) current or past psychiatric disorders (e.g., depression, schizophrenia, anxiety disorders), (2) severe somatic or neurological disorders, (3) history of psychostimulant or other psychotropic medication use, (4) history of alcohol or drug dependence, or familial psychiatric disorders, (5) left-handedness, and (6) contraindications to MRI (e.g., metallic implants).

Forty-two neurotypical children were recruited from a local elementary school using the same exclusion criteria as the ADHD group, matched for age, gender, and handedness. The study was approved by the Ethics Committee of the First Affiliated Hospital of Wenzhou Medical University. All participants and their guardians were fully informed about the study’s purpose and procedures. Written consent was obtained from the guardians, and the relevant informed consent form was signed.

Fifteen children with ADHD and three children with healthy controls (HC) were excluded due to incomplete MRI or behavioral data collection. Eight children with ADHD and three children with HC were excluded due to poor fMRI data quality and excessive head movements. Additionally, outlier analysis was performed on all clinical and neuropsychological measures using the ± 3SD criterion; no participants met this exclusion threshold. Sensitivity analyses confirmed that all significant correlations remained robust after removing extreme values. Finally, 42 children with ADHD and 36 children with HC were retained for statistical analysis. While no formal a priori sample size calculation was performed due to the exploratory nature of this study, post-hoc power analysis indicates that our sample size provides adequate power (> 80%) to detect medium to large effect sizes in between-group comparisons and moderate correlations (r > 0.3) in correlation analyses.

Clinical assessment

All participants were required to complete the IVA-CPT on a computer in a quiet environment to assess auditory attention. Based on our primary research hypothesis regarding auditory attention deficits in ADHD, the primary outcome measures were: (1) auditory attention quotient from the IVA-CPT, and (2) seed-based functional connectivity patterns of core auditory regions (bilateral Heschl’s gyrus, superior temporal gyrus, planum temporale, planum polare, and insula). Secondary outcome measures included visual and combined attention quotients from IVA-CPT, clinical symptom ratings, neuropsychological assessments, and correlational analyses between neural connectivity and behavioral measures.

The Chinese version of the Conners’ Parent Rating Scale (CPRS) [41] was used to assess the severity of clinical symptoms in the subjects, with 48 items across six domains: conduct problems, learning problems, psychosomatic problems, impulsivity-hyperactivity, anxiety, and the hyperactivity index.

Additionally, participants underwent several neuropsychological assessments, including the Chinese Wechsler Intelligence Scale for Children (C-WISC) [42], the Wisconsin Card Sorting Test (WCST) [43], and the Stroop Test [44]. The C-WISC consists of 11 subtests, including 6 verbal subtests (e.g., knowledge, classification, comprehension, arithmetic, vocabulary, and numerical breadth) and 5 performance subtests (e.g., fill-in-the-blank, picture arranging, block patterning, graphic puzzles, and coding). Finally, the Verbal intelligence quotient (VIQ), Performance intelligence quotient (PIQ), and Full-scale intelligence quotient (FIQ) scores were computed [45]. The Stroop Test is administered via a computer program, where participants are instructed to press the left button when they see the color red and the right button when they see blue. The program consists of five test conditions: word-color congruence, word-color conflict, word-color irrelevance, word-color semantic irrelevance with phonological relevance, and neutral stimulation. The system records four raw scores: correct responses, incorrect responses, missed trials, and reaction time. It is an essential tool for assessing selective attention, inhibition of irrelevant stimuli, and impulse control in individuals with ADHD. The WCST requires participants to match cards based on a specific feature (e.g., number, color, or shape) for 10 consecutive trials. Afterward, the system alters the matching strategy. The results are categorized into: total correct responses, correct categories, total errors, persistent errors, and non-persistent errors. This test primarily assesses cognitive flexibility, including perceptual learning, set-shifting, working memory, and executive control.

MRI data acquisition

All participants underwent brain imaging using a 3.0T magnetic resonance scanner with an eight-channel phased-array head coil (signal HDx, General Electric, Milwaukee, WI, USA) to acquire structural and functional images. During scanning, participants were instructed to lie supine, relax with eyes closed to avoid systematic thinking, and have their heads tightly immobilized on a foam pad to minimize movement. Earplugs were used to reduce noise interference.

Whole-brain BOLD data from rs-fMRI were acquired using a gradient-echo echo-planar imaging sequence with the following parameters: number of slices = 31, slice thickness = 4 mm, slice gap = 0.2 mm, and repetition time (TR) = 2000 ms. Repetition time (TR) = 2000 ms, echo time (TE) = 30 ms, flip angle = 90°, matrix size = 64 × 64, and field of view (FOV) = 192 × 192 mm. A total of 240 time points were collected for each subject.

Imaging preprocessing

In this study, we used the Image Processing and Analysis Toolkit (DPABI) on the Matlab platform to preprocess the brain imaging data [46]. The preprocessing steps included: (1) Data format conversion: original DICOM data were converted to NIFTI format. (2) Removal of the first 10 time points of each subject’s rs-fMRI data due to unstable imaging units at the beginning of scanning. (3) Time-layer correction: correction of the acquisition time difference between slices at the same time point. (4) Head movement correction: Excessive head movement can affect image quality and subsequent analysis. Subjects with translational movement greater than 3 mm or angular shift greater than 3.0° during scanning were excluded. The Friston-24 head movement model was used to regress head movement effects, and the mean framewise displacement (mean FD) was calculated. Subjects with mean FD < 0.2 mm were included. (5) Spatial normalization: The realigned images were registered to the Montreal Neurological Institute (MNI) spatial template with a voxel size of 3 × 3 × 3 mm³. (6) Smoothing: Normalized images were smoothed using an 8 mm full-width at half-maximum (FWHM) Gaussian kernel to remove abnormal BOLD values. (7) Elimination of linear drift: Removal of the effects of systematic drift or trends in fMRI data. (8) Filtering: High-frequency noise signals were eliminated, leaving a low-frequency BOLD signal (0.01–0.08 Hz) to reduce the effects of physiological processes such as heartbeat and respiration.

Seed-based functional connectivity analysis

Bilateral PP, HG, PT, and insula were used as seed regions. Eight seed regions were defined using the Harvard-Oxford structural atlas in the DPABI toolbox, and functional connectivity analysis was performed based on these seed regions. The time series for each seed region were obtained by averaging the time series of all voxels within the region. A voxel-by-voxel correlation analysis was then performed between the seed regions and all other voxels across the whole brain to generate functional connectivity maps for each subject. Finally, correlation coefficients were converted to Fisher z values to improve the normality of the functional connectivity maps using Fisher’s r-to-z transformation.

Statistical analysis

Behavioral statistics

SPSS (version 18) was used to calculate between-group differences in demographic and clinical variables. Age, IQ, and neuropsychological assessments (WCST, Stroop test, IVA-CPT) were analyzed using two-sample t-tests. Gender distributions were also compared using the Pearson chi-square test. CPRS scores were analyzed using the Mann-Whitney U test.

Imaging statistics

A two-sample t-test in SPM12 was used to compare the two FC maps between groups, with gender and age as covariates. The statistical threshold was set at voxel-level P < 0.001 (uncorrected), and cluster-level family-wise error (FWE) correction at P < 0.05 with cluster size >100 voxels. The cluster size threshold of 100 voxels was selected following established neuroimaging practices [47, 48], which balances statistical stringency with sensitivity for detecting meaningful connectivity differences while maintaining appropriate control for multiple comparisons. Additionally, a two-sample t-test in SPSS was used to compare head movement levels between the two functional MRI groups.

Correlation analysis

To explore the relationship between group differences in FC values at each auditory seed region and clinical symptoms and attention test results, partial correlation analysis was conducted separately in both ADHD and HC groups, with age and gender as covariates, to examine the relationship between FC values in brain regions showing significant between-group differences and neuropsychological measures. Comparative analysis revealed that all significant brain-behavior correlations were specific to the ADHD group, with no corresponding associations observed in HC, indicating these represent ADHD-specific pathological relationships rather than general developmental patterns (digit span on the C-WISC, correct, incorrect, missed responses, and reaction times on the Stroop test, and correct/incorrect classifications and persistent/non-persistent errors on the WCST), attention tests (Auditory, Visual, and Combined Attention Quotients from the IVA-CPT), and clinical symptom scores (the six dimensions of the CPRS). A statistical threshold of p < 0.05 was set for correlation analyses. Following established practices in exploratory brain-behavior studies [49], no correction for multiple comparisons was applied to preserve sensitivity for detecting meaningful associations in this hypothesis-generating study. Effect sizes are reported alongside p-values to assess practical significance of observed correlations.

Result

Demographic and clinical characteristics of ADHD and HC

No significant differences were found between the ADHD and HC groups in age (p = 0.748), gender (p = 0.092), IQ scores (verbal: p = 0.187; performance: p = 0.487; full-scale: p = 0.244), or head movement (p = 0.238). Significant group differences were observed in CPRS clinical symptom scores, with the ADHD group showing higher scores in conduct problems (p < 0.001), learning problems (p < 0.001), impulsivity-hyperactivity (p < 0.001), and hyperactivity index (p < 0.001). The ADHD group also demonstrated longer reaction times in the Stroop test (p = 0.003) and significantly lower IVA-CPT attention quotients: auditory (p < 0.001), visual (p < 0.001), and combined (p < 0.001), confirming significant attention deficits as hypothesized. Specific demographic and test results are presented in Table 1. Further results from the two-sample t-test analysis of resting-state functional connectivity (rsFC) are provided in Tables 2 and 3; Fig. 1.

Table 1.

Demographic and clinical characteristics and between-group comparisons of ADHD and healthy control participants

	ADHD(n = 42)	HC(n = 36)	T/Z value	p-value
Age	9.26 ± 2.51	9.42 ± 1.70	-0.32	0.748a
Gender(M/F)	32/10	21/15	2.838	0.092b
Verbal IQ	123.88 ± 18.91	129.38 ± 17.41	-1.33	0.187a
Performance IQ	105.93 ± 15.26	108.28 ± 14.24	-0.7	0.487a
Full-scale IQ	117.79 ± 17.10	121.94 ± 13.65	-1.17	0.244a
Head Motion	0.079 ± 0.031	0.091 ± 0.053	-1.19	0.238a
Stroop Test
Reaction Time	3193.50 ± 479.13	2466.47 ± 1292.50	3.19	0.003a
Correct Responses	47.76 ± 17.94	52.33 ± 19.36	-1.08	0.283a
Error Responses	35.76 ± 10.68	32.36 ± 11.25	1.37	0.176a
Omission Errors	38.42 ± 19.12	36.16 ± 20.98	0.5	0.62a
CPRS
Conduct problems	1.04 ± 0.54	0.38 ± 0.32	-5.43	< 0.001c
Psychosomatic problems	0.30 ± 0.38	0.15 ± 0.23	-1.73	0.083c
Anxiety	0.41 ± 0.32	0.47 ± 0.29	-1.31	0.896c
Learning problems	1.70 ± 0.59	0.68 ± 0.70	-5.65	< 0.001c
Impulsivity–hyperactivity	1.44 ± 0.68	0.46 ± 0.50	-5.79	< 0.001c
Hyperactivity index	1.36 ± 0.55	0.48 ± 0.41	-6.28	< 0.001c
WCST
Total Number of Correct Responses	20.81 ± 7.46	21.81 ± 7.45	-0.59	0.558a
Total Number of Errors	27.19 ± 7.46	26.19 ± 7.45	0.59	0.558a
Perseverative Errors	12.71 ± 7.71	11.86 ± 5.21	0.56	0.576a
Nonperseverative Errors	14.48 ± 6.36	14.33 ± 4.77	0.11	0.912a
Number of Categories Completed	2.20 ± 1.19	2.25 ± 1.16	-0.22	0.824a
IVA-CPT
Full scale control quotient	80.48 ± 22.56	79.64 ± 22.80	0.163	0.871a
Visual control quotient	80.36 ± 27.16	82.78 ± 23.07	-0.42	0.675a
Auditory control quotient	82.12 ± 22.72	79.50 ± 23.23	0.5	0.617a
Full scale attention quotient	72.29 ± 22.42	95.61 ± 18.64	-4.95	< 0.001a
Visual attention quotient	71.17 ± 26.36	93.67 ± 19.39	-4.23	< 0.001a
Auditory attention quotient	73.93 ± 23.94	96.61 ± 18.20	-4.75	< 0.001a

ADHD, Attention Deficit Hyperactivity Disorder; HC, health control; CPRS, Conners’ Parent Rating Scale; WCST, Wisconsin Card Sorting Test; IVA-CPT, Auditory Continuous Performance Test; M, male; F, female; T/Z values represent statistical comparisons between ADHD and HC groups, with t-tests used for continuous variables and chi-square/Mann-Whitney U tests used for categorical and non-parametric variables respectively

^ap values were obtained using two-sample t tests

^bp values were obtained using a chi-square test

^cp values were obtained using Mann–Whitney U test

Table 2.

Region of decreased RSFC in ADHD compared with the HC

Seed regions	Target areas	Cluster size	Peak MNI coordinate			t value	pFWE value
			x	y	z
Left Insular	Left cerebellum	422	-27	-60	-54	6.32	< 0.001
Left HG	Left superior temporal gyrus	201	-45	-39	9	7.19	0.02
Left PT	Right insula	172	42	3	0	4.51	0.032
Right Insular	Left cerebellum	578	-27	-63	-51	5.19	< 0.001
	Right cerebellum	231	15	-39	-30	5.2	0.008
	Left lingual gyrus	136	-9	-78	0	3.81	0.049
	Left superior temporal gyrus	156	-45	-45	15	5.32	0.033
Right PP	Left cerebellum	434	-27	-69	-51	4.92	< 0.001
	Right cerebellum	200	33	-42	-30	4.22	0.015
Right PT	Right insula	197	39	0	6	4.85	0.017

ADHD, Attention Deficit Hyperactivity Disorder; HC, health control; HG, Heschl’s gyrus; PT, planum temporale; PP, planum polare; pFWE, p value with FWE correction

Table 3.

Region of increased FC in ADHD compared with the HC

Seed regions	Target areas	Cluster size	Peak MNI coordinate			t value	pFWE value
			x	y	z
Right HG	Right superior frontal gyrus	171	21	18	54	4.31	0.025
Right PT	Right supramarginal gyrus	456	54	-42	36	4.77	< 0.001
	Right middle frontal gyrus	367	27	21	54	4.86	0.001
	Left middle frontal gyrus	148	-30	15	48	4.57	0.042

ADHD, Attention Deficit Hyperactivity Disorder; HC, health control; HG, Heschl’s gyrus; PT, planum temporale; pFWE, p value with FWE correction

Fig. 1.

(A) Seed regions in the auditory cortex and associated areas. (B) Regions showing decreased rsFC in ADHD compared with HC (HC > ADHD). (C) Regions showing increased rsFC in ADHD compared with HC (ADHD > HC). Color bars represent t-values from two-sample t-tests. Results are displayed at voxel-level P < 0.001 (uncorrected), cluster-level FWE-corrected P < 0.05. Coordinates are in MNI space (mm). ADHD, attention deficit/hyperactivity disorder; HC, healthy control; PP, planum polare; HG, Heschl’s gyrus; PT, planum temporale; L, left; R, right

Seed regions FC

Planum polare FC

Compared to HC, children with ADHD exhibited significantly decreased rsFC between the right PP and bilateral cerebellum (left cerebellum: P < 0.001, right cerebellum: P = 0.015, FWE-corrected). The rsFC of the left PP showed no significant group differences.

Heschl’s gyrus FC

The left and right HGs showed opposite results: a significant decrease in rsFC between the left HG and the left STG in the ADHD group (P = 0.020, FWE-corrected) and a significant increase in rsFC between the right HG and the right superior frontal gyrus (SFG) (P = 0.025, FWE-corrected).

Planum temporale FC

In the ADHD group, the rsFC between the bilateral PT and the right insula was significantly decreased (left PT: P = 0.032, right PT: P = 0.017, FWE-corrected), while rsFC between the right PT and both the right supramarginal gyrus (SMG) and middle frontal gyrus (MFG) was significantly increased (right SMG: P < 0.001, left MFG: P = 0.042, right MFG: P < 0.001, FWE-corrected). No significant increases were found in the left PT.

Insula FC

RsFC between the left insula and the left cerebellum was significantly decreased compared to HC (P < 0.001, FWE-corrected). The right insula showed decreased rsFC in additional brain regions, including the bilateral cerebellum (left cerebellum: P < 0.001, right cerebellum: P = 0.049, FWE-corrected), the left lingual gyrus (P = 0.044, FWE-corrected), and the left STG (P = 0.033, FWE-corrected).

Correlation analysis results

Correlation analyses were conducted between all identified functional connectivity alterations and all neuropsychological measures. Significant correlations in the ADHD group included: rsFC between the left HG and left STG was negatively correlated with the auditory attention quotient on the IVA-CPT. Complete correlation matrices are available from the corresponding author upon request. Additionally, rsFC between the right PT and left MFG was negatively correlated with the psychosomatic problem score in the ADHD group. rsFC between the right PP and left cerebellum, as well as between the right PT and right insula, were positively correlated with digit span on the C-WISC in the ADHD group. Furthermore, rsFC between the right PT and right MFG was positively correlated with the number of non-sustained errors on the WCST in the ADHD group. The results of these correlations are presented in Fig. 2.

Fig. 2.

The results of partial between correlation seed-based functional connectivity alterations and clinical measures. Y-axis represents functional connectivity values as Fisher z-transformed correlation coefficients, with higher values indicating stronger positive connectivity and lower values indicating weaker or negative connectivity. ADHD, attention deficit/hyperactivity disorder. PP, planum polare. HG, Heschl’s gyrus. PT, planum temporale. L, left. R, right. MFG, middle frontal gyrus. STG, superior temporal gyrus

Discussion

This study investigated auditory attention and the functional connectivity of the auditory cortex with the whole brain in children with first-onset, unmedicated ADHD. Compared to the HC group, children with ADHD exhibited significantly poorer auditory attention. We further examined functional connectivity using the auditory cortex and other auditory-related regions as seed regions and found significant decreased rsFC between bilateral insula and cerebellum, as well as between the right PP and cerebellum. Additionally, compared to HC, children with ADHD showed decreased connectivity between the left HG and the left STG, between the right insula and the left STG, as well as between bilateral PT and the right insula. Compared to HC, children with ADHD showed a significant increase in functional connectivity between the right HG and the right SFG, as well as between the right PT and the right SMG and the MFG bilaterally. Further correlation analysis indicated that the abnormal functional connectivity between the left Heschl’s gyrus and the left superior temporal gyrus was associated with the auditory attention quotient. These findings support our hypothesis that children with ADHD exhibit auditory attention deficits, which are linked to abnormal functional connectivity in the auditory cortex.

Auditory attention deficits in children with ADHD

We assessed auditory attention in children with ADHD using the IVA-CPT, and our findings revealed deficits in maintaining attention to auditory stimuli, consistent with previous studies. Similar studies using the IVA-CPT have reported significant differences in auditory attention between ADHD and control groups [19, 37, 50]. Wu et al. [19] demonstrated that the IVA-CPT effectively differentiates ADHD children from controls through comprehensive auditory attention measures, while Moreno-García et al. [37] confirmed significant auditory attention impairments in ADHD children using the IVA-CPT methodology. Additionally, Thompson et al. [50] reported consistent findings of auditory attention deficits in ADHD populations using similar continuous performance testing approaches. However, our results did not show a statistically significant difference in the auditory control quotient between children with ADHD and HC (p = 0.617), consistent with some studies [37]. In contrast, Wang et al. [39] found that children with ADHD not only had deficits in auditory attention but also significant differences in auditory control. This dissociation between preserved auditory control and impaired auditory attention in our ADHD sample suggests that basic auditory processing abilities remain intact while sustained attention mechanisms are specifically compromised. Importantly, when auditory control quotient was included as a covariate in our brain-behavior correlation analyses, all significant associations remained robust, indicating that FC alterations reflect attention-specific neural mechanisms rather than general auditory processing deficits or potential confounding by basic auditory control abilities. Further detailed analyses using the IVA-CPT suggested that children with ADHD experience difficulties at both the auditory sensorimotor and auditory comprehension levels [37]. In summary, we have confirmed that children with ADHD exhibit auditory attention deficits, as demonstrated by the IVA-CPT, and further evidenced abnormalities in their auditory processing.

Regions with decreased resting-state functional connectivity in ADHD patients

Compared to HC, we identified several regions in children with ADHD exhibiting decreased functional connectivity to the auditory cortex and other auditory-related areas, including the cerebellum, left superior temporal gyrus, and right insula. Early studies emphasized the cerebellum’s role in motor control [51], but subsequent research has revealed its impact on language [50], cognition [52], and emotion [53, 54], as well as its involvement in executive function and attention deficits due to cerebellar damage [55]. Neuroimaging studies support this conclusion, showing significant structural and functional abnormalities in the cerebellum of ADHD patients, including reduced cerebellar size in children with ADHD compared to controls [56], which correlates with behavioral outcomes such as hyperactivity, inattention, and impulsivity, as reported by parents [57]. Petacchi et al. [58] used fMRI and positron emission tomography (PET) techniques to demonstrate that specific regions of the cerebellum are activated during auditory tasks and contribute to auditory sensory processing. Our study found abnormal functional connectivity between the cerebellum and several auditory-related brain regions, further supporting the cerebellum’s critical role in auditory processing. Our results showed decreased functional connectivity between the cerebellum and the bilateral insula, as well as with the right PP. Poor coordination among the networks of these regions may underlie Auditory attention deficits in children with ADHD. These connectivity reductions likely represent primary pathological disruptions in core auditory processing circuits, directly contributing to the attention deficits characteristic of ADHD.

Additionally, our results revealed a significant reduction in functional connectivity between the left HG and left STG. The superior temporal gyrus, located in the temporal lobe near the lateral surface, is a core region of the auditory cortex [59, 60]. The Heschl’s gyrus is the core of the primary auditory cortex, involved in the initial processing of sound, particularly in resolving frequency, pitch, and temporal features crucial for both auditory processing and language [61]. Thus, the functional connectivity between the HG and STG plays a crucial role in regulating auditory processing and contributes to typical left-hemisphere language dominance. The reduced left HG-STG connectivity in ADHD may reflect disrupted language lateralization patterns commonly reported in neurodevelopmental disorders. Recent studies have shown that the volume of the left STG is reduced in children with ADHD compared to typically developing children. This volume is positively correlated with working memory capacity [62], which is linked to cognitive control, and reduced working memory often results in auditory attention deficits [25]. Increased HG-STG connectivity has also been reported in adults with ADHD, contrasting with our study, this study investigated multisensory system integration in adult ADHD patients using diffusion MRI [26]. This study, conducted on adult ADHD patients, suggests that age may play a role in the functional connectivity between the HG and STG. Children’s brain development is still incomplete, particularly in those with ADHD, where brain regions lag behind and the auditory cortex and language network remain immature. This leads to decreased functional connectivity between the HG and STG, impairing auditory processing and contributing to auditory attention deficits.

The insula, particularly the posterior insula, plays a key role in processing and integrating information from various sensory modalities, including auditory and tactile stimuli [63]. Neuroimaging studies have shown that the insula is involved in modulating auditory attention, with resting-state functional connectivity of the right insula being significantly altered in children with auditory processing deficits [64]. Our findings also revealed abnormal functional connectivity between the right insula and the PT, cerebellum, lingual lobe, and STG, further suggesting that the insula is crucial to the neural mechanisms underlying auditory attention deficits in children with ADHD.

Regions with increased resting-state functional connectivity in ADHD patients

Compared to HC, our results demonstrate that children with ADHD show a significant increase in functional connectivity between auditory-related regions and the right SMG and frontal lobes, including the superior and middle frontal gyri. These connectivity increases may reflect compensatory recruitment of higher-order attentional control networks attempting to maintain performance despite primary circuit dysfunction, though this compensation appears insufficient as evidenced by persistent behavioral deficits. The supramarginal gyrus is crucial for perceptual and attentional control, particularly in processing external stimuli (e.g., visual, tactile, auditory), and plays a key role in auditory information processing [65, 66]. The frontal lobes are integral to the cognitive and behavioral features of ADHD, including inhibitory control, attention regulation, and working memory [67]. The supramarginal gyrus and superior frontal gyrus exhibit complex network affiliations, with connectivity patterns that can overlap across multiple brain networks depending on task demands and individual differences [68]. While the supramarginal gyrus is primarily associated with ventral attention and salience networks, it also shows connectivity with DMN components, particularly in the context of attention regulation [69]. Similarly, the superior frontal gyrus, particularly its medial aspects, can function as part of cognitive control networks while also exhibiting DMN-related connectivity patterns [70, 71]. Our seed-based functional connectivity analysis revealed increased connectivity between auditory regions and these frontal/parietal areas in children with ADHD. While our methodology did not directly examine DMN through canonical DMN seeds, these findings suggest altered interactions between auditory processing regions and higher-order cognitive networks that may include DMN components. The increased connectivity patterns may reflect disrupted network segregation and altered attention regulation mechanisms characteristic of ADHD, though the specific contributions of DMN versus attention/control networks require further investigation using network-specific analytical approaches. Blomberg et al. [27] also found a stronger correlation between the auditory network and the right supramarginal gyrus, with enhanced functional connectivity, which may explain why ADHD patients are more easily distracted during auditory tasks. Using magnetoencephalography, Heinrichs-Graham et al. [72] found that adults with ADHD had poorer auditory attention task performance, with increased recognition accuracy associated with enhanced functional connectivity between the dorsolateral prefrontal cortex (DLPFC) and the auditory cortex (Heschl’s gyrus). Two other studies have also found significantly increased connectivity between DMN regions and auditory cortex areas involved in auditory processing in ADHD subjects [33, 34]. In summary, we conclude that children with ADHD exhibit abnormalities in functional connectivity between auditory regions and brain networks, such as the frontal lobes, which affect auditory processing and top-down attentional modulation.

Correlation analysis

We found that the rsFC between the left HG and the left STG was negatively correlated with the auditory attention quotient on the CPT in the ADHD group. The auditory attention quotient, based on an individual’s response to auditory stimuli, assesses the concentration and stability of attention to auditory information, with higher scores indicating better attention maintenance. This correlation suggests that abnormal functional connectivity in the primary auditory cortex (HG) is closely linked to auditory attention deficits. Previous studies have reported significant structural and functional abnormalities in the Hirschsprung’s gyrus in ADHD patients. Our results provide additional evidence for the underlying mechanisms of auditory attention deficit in children with ADHD.

Additionally, we found that the rsFC between the right PP and the left cerebellum, as well as between the right PT and the right insula, were positively correlated with digit span on the Wechsler Intelligence Test in the ADHD group. Digit span tests are commonly used to assess working memory, a key component of executive functioning, which is known to be impaired in ADHD patients [73]. Our results highlight the relationship between working memory and auditory processing. Previous studies suggest that working memory influences auditory attention in ADHD patients, with individuals possessing higher working memory capacity better able to maintain attention during tasks compared to those with lower capacity. Similarly, the rsFC between the right PT and the right MFG was positively correlated with the number of non-sustained errors on the Wisconsin Card Sorting Test in the ADHD group. The number of non-sustained errors on the WCST was also correlated with executive functioning [74], with lower scores indicating poorer executive function in individuals with ADHD, further illustrating the impact of the executive function deficit.

Finally, we found that the rsFC between the right PT and the left MFG was negatively correlated with the psychosomatic scores in the ADHD group. Higher psychosomatic scores indicate more severe psychosomatic problems in ADHD. Previous studies have shown that children with ADHD often experience more psychosomatic issues. Our findings suggest that these issues may be linked to abnormal functional connectivity in specific brain regions.

Limitation

Although this study supports our previous hypotheses, several limitations must be considered. First, there is a lack of a standardized anatomical or functional template of the brain tailored for children. The use of adult-derived neuroanatomical templates (Harvard-Oxford Atlas) for children seed region selection introduces developmental validity concerns. Second, our sample exhibits a gender distribution imbalance (ADHD: 76.2% male; HC: 58.3% male). While we statistically controlled for gender as a covariate, this imbalance may limit generalizability given established sex differences in functional connectivity patterns. Third, the insula is a large brain region involved in various cognitive processes. It not anatomically segregated in standard atlases, potentially obscuring circuit-specific connectivity alterations. Fourth, our study focused on children with first-onset, unmedicated ADHD, controlling for heterogeneity. However, further research is needed to determine whether these findings apply to medicated children or other age groups with ADHD. Finally, ADHD has a complex etiology and is typically classified into three subtypes. Due to sample size limitations, subtype analysis was not performed in this study, and future research with larger samples is needed.

Conclusion

This study confirmed auditory attention deficits in children with ADHD and identified abnormal functional connectivity in the auditory cortex and related regions. Specifically, the rsFC between the left HG and left STG was negatively correlated with auditory attention levels in the ADHD group. These findings suggest that auditory attention deficits in children with ADHD are linked to abnormal auditory processing in the auditory cortex and related areas. Finally, our findings relate to the presence of abnormal brain networks in ADHD found in previous studies, such as the DMN and SN, which also influence sensory processing. Overall, our findings provide neuroimaging evidence of auditory attention deficits in children with ADHD. These abnormal regions could serve as potential therapeutic targets for clinical interventions aimed at ADHD children with auditory attention deficits.

Abbreviations

ADHD

Attention Deficit/Hyperactivity Disorder

FC

Functional connectivity

HC

Healthy controls

IVA-CPT

Integrated Visual and Auditory Continuous Performance Test

rs-fMRI

Resting-state functional magnetic resonance imaging

MRI

Magnetic resonance imaging

HG

Heschl’s gyrus

STG

Superior temporal gyrus

PP

Planum polare

PT

Planum temporale

CPRS

Conners’ Parent Rating Scale

C-WISC

Chinese Wechsler Intelligence Scale for Children

WCST

Wisconsin Card Sorting Test

VIQ

Verbal intelligence quotient

PIQ

Performance intelligence quotient

FIQ

Full-scale intelligence quotient

rsFC

Resting-state functional connectivity

SFG

Superior frontal gyrus

SMG

Supramarginal gyrus

MFG

Middle frontal gyrus

DMN

Default Mode Network

VAN

Ventral attention network

SN

Salience network

DLPFC

Dorsolateral prefrontal cortex

Author contributions

MZ, JY and CY designed the study, contributed to data analysis, MZ and JY interpreted the data and wrote the manuscript; CY contributed to study supervision, obtained funding, and reviewed and commented on the first draft of the manuscript; RJ, JZ, XC and TZ contributed to data acquisition and processing; HL and WZ assisted with data analysis and interpretation of findings; XH, MW and CY viewed and revised the manuscript.MZ and JY Contributed equally in the study.All authors critically reviewed and approved the final version of the manuscript submitted for publication.

Funding

This study was supported by grants from the Department of Science and Technology of Zhejiang Province (Grant No. Y2024C04021).

Data availability

The datasets generated and/or analyzed during the current study are not publicly available due to privacy and ethical restrictions but are available from the corresponding author on reasonable request.

Declarations

Ethical approval and consent to participate

The study was conducted in accordance with the Declaration of Helsinki and was reviewed and approved by the Ethics Committee of the First Affiliated Hospital of Wenzhou Medical University. Because the participants were minor children, written informed consent to participate in this study was provided by the participant’s legal guardian.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.