Altered Resting-State Neural Oscillations and Functional Connectivity in Persistent Postural-Perceptual Dizziness

1. Introduction

Persistent postural-perceptual dizziness (PPPD) is a chronic functional neuro-otologic disorder characterized by persistent dizziness, non-spinning vertigo, or unsteadiness lasting for at least 3 months and exacerbated by upright posture, active or passive motion, and exposure to moving or complex visual stimuli [1]. It is increasingly recognized as a common cause of chronic vestibular symptoms in neurology and dizziness clinics, and may impose a substantial burden on daily activities, emotional well-being, and health-related quality of life [2,3,4]. Despite standardized diagnostic criteria, PPPD remains clinically challenging because vestibular tests and conventional neuroimaging are often non-specific, whereas symptoms are frequently accompanied by anxiety, depression, visual dependence, and impaired postural control [5,6]. Consequently, objective neurophysiological markers that reflect the underlying network dysfunction are still needed.

Current evidence suggests that PPPD is not solely attributable to peripheral vestibular dysfunction, but may involve maladaptive interactions among vestibular, visual, somatosensory, postural, and affective systems [7,8]. Structural and functional neuroimaging studies have reported abnormalities in multimodal vestibular regions, visual cortices, insular networks, frontal regulatory regions, and cerebellar–sensorimotor circuits [9,10,11,12]. Network-level alterations have also been described in patients with phobic postural vertigo and PPPD, supporting the view that chronic dizziness may be associated with altered cortical integration rather than isolated vestibular end-organ dysfunction [13]. However, most available studies have relied on MRI-based measures, which provide limited temporal resolution for characterizing neural oscillatory dynamics. Although clinical and neurophysiological studies have begun to explore electrophysiological features in PPPD, resting-state EEG evidence remains limited, particularly regarding frequency-specific power changes and phase-based functional connectivity patterns [14]. In addition, few studies have integrated EEG indices with quantitative posturography in the same cohort [15].

The present study was designed to investigate resting-state EEG abnormalities in patients with PPPD using spectral power analysis and weighted phase lag index-based functional connectivity. We hypothesize that PPPD would be associated with frequency-specific alterations in cortical oscillatory activity and functional connectivity involving prefrontal, visual, temporal, cerebellar, and sensorimotor regions. We further propose that these EEG alterations would be related to dizziness-related disability and postural instability. To test these hypotheses, patients with PPPD and age- and sex-matched healthy controls were assessed using clinical scales, static posturography, and 71-channel resting-state EEG. This study aimed to clarify whether resting-state electrophysiological markers may help characterize the neural network alterations associated with PPPD and provide a basis for future mechanistic and longitudinal investigations.

2. Materials and Methods

2.1. Participants

This study enrolled 40 patients with persistent postural-perceptual dizziness (PPPD) who were recruited from the outpatient and inpatient services of Nanjing Brain Hospital between April 2025 and April 2026. The PPPD group included 15 men and 25 women, with a mean age of 58.58 ± 9.76 years. All patients met the 2017 Bárány Society diagnostic criteria for PPPD. The inclusion criteria were as follows: (1) age between 30 and 75 years; (2) absence of other neurological disorders, including traumatic brain injury, intracranial tumors, or cerebrovascular disease; (3) no use of medications that could affect resting-state brain activity, such as antidepressants or anxiolytics, within 4 weeks before enrollment.

Forty age- and sex-matched healthy controls were recruited. Control participants had no history of otologic, neurological, or psychiatric disorders and had not used medications known to affect resting-state brain activity. The study protocol was approved by the Ethics Committee of Nanjing Medical University Affiliated Brain Hospital(Approval No: IRB-AF40-1.0). Written informed consent was obtained from all participants before study participation.

2.2. Clinical Assessment

Demographic and clinical data were collected before EEG acquisition. The recorded variables included sex, age, and disease duration. Dizziness severity and balance-related symptoms were assessed using the Visual Analog Scale (VAS), Dizziness Handicap Inventory (DHI) [16,17], and Activities-specific Balance Confidence Scale (ABC) [18]. Psychologic symptoms were evaluated using the Hamilton Anxiety Rating Scale (HAMA) and Hamilton Depression Rating Scale (HAMD).

2.3. Static Posturography

Static postural stability was assessed using the Pro-Kin visual feedback balance system (PK254; TecnoBody S.R.L., Bergamo, Italy) [19,20]. Participants were instructed to stand comfortably on the force platform in a standardized position, with their arms placed alongside the body and their gaze fixed on a target located directly in front of them. Each participant completed two 30-second trials under eyes-open (EO) and eyes-closed (EC) conditions. Center-of-pressure (COP) data were recorded using pressure-sensitive sensors. The following posturographic parameters were analyzed: COP trajectory length, ellipse sway area, and mean sway velocity in the anterior–posterior and medio-lateral directions. Greater trajectory length, larger sway area, and higher sway velocity were interpreted as indicators of poorer postural stability. See Figure 1 for further details on static posturography.

2.4. EEG Acquisition

Resting-state EEG was recorded using a 71-channel whole-head EEG cap system (BoruiKang Technologies, Changzhou, China) [21]. The electrode montage covered conventional cortical regions and cerebellar electrodes, including PO9, PO10, O9, O10, Iz, CBz, CB1, and CB2. Electrode placement followed the international 10–10 system. The reference electrode was positioned at CPz, and the ground electrode was positioned at AFz.

EEG signals were sampled at 1000 Hz. Electrode impedance was maintained below 20 kΩ throughout acquisition. Recordings were performed during the daytime in a quiet environment. Participants were asked to wash their scalp before testing and were screened to ensure the absence of hunger, hypoglycemia, fatigue, or other discomfort. During recording, participants sat quietly with their eyes closed, remained awake, relaxed their facial and body muscles, and minimized horizontal and vertical eye movements. A 10-minute eyes-closed resting-state EEG recording was obtained from each participant.

2.5. Spectral Power Analysis

Spectral power analysis was performed in MATLAB [22]. The Welch method was applied to resting-state EEG data using a Hamming window, with 50% overlap between adjacent segments. The segment length was set to one-eighth of the total data length. Absolute power was calculated for the following frequency bands: delta, 1–4 Hz; theta, 4–8 Hz; alpha, 8–13 Hz; beta, 13–30 Hz; and gamma, 30–50 Hz. Relative power was then calculated for each frequency band. Cluster-based permutation testing was used to assess between-group differences in spectral power while controlling for multiple comparisons across electrodes.

2.6. Functional Connectivity Analysis

Functional connectivity was assessed using the weighted phase lag index (wPLI), which estimates phase synchronization between EEG signals while reducing the influence of volume conduction [23,24]. For each participant, 67 × 67 functional connectivity matrices were constructed for the delta, theta, alpha, beta, and gamma frequency bands. Mean wPLI values were then calculated within each frequency band and compared between groups.

2.7. Statistical Analysis

Statistical analyses were performed using SPSS version 25.0 (SPSS Inc., Chicago, IL, USA) and MATLAB. Normality and homogeneity of variance were examined before group comparisons. Continuous clinical variables were compared using independent-samples t tests when parametric assumptions were satisfied. For EEG spectral power and functional connectivity analyses, between-group differences were assessed across frequency bands. Cluster-based permutation statistics were applied to spectral power analyses to control for multiple comparisons across EEG electrodes. For Functional connectivity, Network-Based Statistics (NBS) were applied to compare wPLI matrices between PPPD groups and healthy controls [25]. To control for potential confounding effects of affective symptoms, group comparisons of EEG spectral power and functional connectivity were reanalyzed using analysis of covariance (ANCOVA), with HAMA and HAMD scores as covariates. Within the PPPD group, correlations between EEG indices and clinical, posturographic, and psychological variables were evaluated using Pearson or Spearman correlation analysis, depending on data distribution. All tests were two-sided, and statistical significance was set at P < 0.05.

3. Results

3.1. Clinical Characteristics

Demographic and clinical characteristics of the subjects are summarized in Table 1. No significant between-group differences were observed in age and sex distribution. Compared with healthy controls, patients with PPPD had significantly higher HAMA and HAMD scores, indicating greater anxiety and depressive symptom burden.

3.2. Static Postural Stability

Static posturography showed impaired postural stability in patients with PPPD compared with healthy controls. Detailed statistical analyses of all parameters were shown in Table 2. Under the eyes-open condition, patients with PPPD showed significantly greater medio-lateral COP displacement (EO-x, P = 0.0013), anterior–posterior COP displacement (EO-y, P = 0.0058), and sway area (EO-area, P = 0.0242). COP trajectory length under the eyes-open condition showed an increasing trend but did not reach statistical significance (EO-length, P = 0.0558).

Under the eyes-closed condition, patients with PPPD showed significantly greater medio-lateral displacement (EC-x, P = 0.0201), anterior–posterior displacement (EC-y, P = 0.0011), sway area (EC-area, P = 0.0070), and trajectory length (EC-length, P = 0.0199) than healthy controls. These findings indicate that postural control was impaired in PPPD and became more unstable when visual input was removed.

3.3. Resting-State EEG Spectral Power

Cluster-based permutation analysis revealed significant between-group differences in theta-band relative power(Figure 2). Compared with healthy controls, patients with PPPD showed increased theta-band power over the frontal region (electrodes :AF3,F1,F3) and right parieto-occipital region (electrodes:PO7,PO9), after controlling for psychiatric comorbidities. No significant between-group differences were detected in delta, alpha, beta, or gamma power.

3.4. Functional Connectivity

wPLI-based functional connectivity analysis showed frequency-specific alteration in patients with PPPD compare to healthy controls (Figure 3). After controlling for psychiatric comorbidities, temporal–occipital connectivity was decreased, whereas prefrontal–temporal connectivity was increased in the theta band. Connectivity among frontal, parietal, and occipital regions was increased, and parietal -temporal connectivity was decreased in the alpha band. Cerebellar and temporal, central connectivity was decreased in the gamma band. No significant differences were observed in cerebellar-related connectivity in the remaining frequency bands.

3.5. Correlation Analysis

Correlation analyses were performed within the PPPD group (Figure 4). Frontal spectral power was positively correlated with HAMA scores (r = 0.332, P =0.036). DHI scores were negatively correlated with ABC scores (r = −0.441, P =0.004). In addition, frontal spectral power was significant correlated with occipital spectral power (r = 0.930, P < 0.001). Age, disease duration, dizziness or balance assessment indices, and EEG spectral power showed no significant correlations in the remaining analyses.

4. Discussion

In this study, PPPD was associated with frequency-specific resting-state EEG alterations, including increased theta-band relative power over frontal and right parieto-occipital regions, altered theta-, alpha-, and gamma-band functional connectivity, impaired postural stability, and a positive association between frontal spectral power and anxiety severity. These findings support the presence of altered electrophysiological patterns involving cognitive-affective, visual, and sensorimotor systems in PPPD, while requiring confirmation in larger longitudinal cohorts.

Theta oscillations have been implicated in long-range communication involved in cognitive control and contextual integration, particularly in fronto-temporal and fronto-parietal systems [26]. The increased theta-band activity over frontal regions and increased prefrontal–temporal connectivity may represent an electrophysiological correlate of altered attentional control, body vigilance, and affective modulation rather than a disease-specific biomarker. This interpretation is consistent with recent conceptualizations of PPPD as a functional neuro-otologic disorder in which persistent symptoms are maintained by maladaptive interactions among postural control, spatial orientation, sensory reweighting, and threat-related processes [5,6]. The positive association between frontal spectral power and HAMA scores further suggests that frontal oscillatory activity may be linked to anxiety burden in PPPD. Affective symptoms, particularly anxiety and depression, are highly prevalent in patients with PPPD and are increasingly recognized as important modulators of both symptom severity and central nervous system functioning. Recent clinical studies have shown that anxiety-related hypervigilance, fear of movement, and maladaptive threat monitoring may contribute to persistent dizziness and postural instability in PPPD, forming part of a self-reinforcing perceptual–affective loop rather than representing purely comorbid psychiatric conditions [27,28,29,30]. In the present study, ANCOVA results demonstrated that after controlling for HAMA and HAMD scores, several EEG abnormalities—particularly posterior theta power and specific theta/alpha/gamma connectivity alterations remained significant. This supports the notion that emotional symptoms may act as modulators rather than sole drivers of PPPD-related neural dysfunction.

The occipital and right parieto-occipital theta-band findings are also consistent with previous neuroimaging evidence implicating visual and posterior cortical regions in PPPD. Li et al. reported altered spontaneous activity in the right precuneus and cuneus in patients with PPPD, indicating that posterior cortical regions involved in visuospatial processing may contribute to the pathophysiology of this disorder [10]. In the present study, increased theta power in the occipital/parieto-occipital region and decreased temporal–occipital connectivity may indicate altered top-down modulation of visual processing during rest. This interpretation is also consistent with studies of visually induced dizziness, in which altered functional brain connectivity has been observed in networks related to visual motion perception and multisensory integration [31,32]. Nevertheless, because EEG source localization was not performed, the anatomical interpretation of scalp-level occipital and parieto-occipital signals should remain cautious.

The decreased theta-band temporal–occipital connectivity and alpha-band parietal-temporal connectivity may reflect impaired integration between vestibular-related, auditory/temporal, visual, and higher-order regulatory systems [33,34]. Previous fMRI work in chronic subjective dizziness demonstrated altered activity in vestibular and insular systems during vestibular stimulation, suggesting that abnormal processing within vestibular–interoceptive networks may contribute to chronic dizziness symptoms [35]. Moreover, Li et al. reported altered intra- and inter-network functional connectivity in PPPD, including abnormal interactions between visual, sensorimotor, and default-mode-related networks [9].The present EEG findings extend these observations by showing that such network alterations may be frequency-specific and detectable in resting-state electrophysiological connectivity.

The increased fronto-parieto-occipital connectivity in alpha-band may be related to enhanced visual–spatial monitoring or increased reliance on visual information for postural control [32,36,37]. This interpretation is supported by the posturographic findings: patients with PPPD showed significantly greater COP displacement and sway area than controls, and instability became more evident when visual input was removed under eye-closed conditions. Similar behavioral observations have been reported in patients with phobic postural vertigo, in whom gait and postural control were influenced by fear of falling, attention, and visual input [38]. Experimental work has also shown that visual motion can modulate postural sway, supporting the role of visual input in balance control [39]. Together with the present EEG findings, these data suggest that excessive visual weighting and altered cortical control of sensory integration may participate in PPPD-related postural instability. However, the present study did not directly manipulate visual motion or vestibular stimulation; therefore, visual dependence was inferred from postural performance rather than experimentally tested.

In the gamma band, decreased connectivity between cerebellar and temporal, central regions may be clinically relevant because cerebellar and sensorimotor networks are important for postural control, movement calibration, and vestibular–motor integration. Prior multimodal imaging work in phobic postural vertigo reported cortical and cerebellar alterations, suggesting that distributed motor and vestibular networks may be involved in functional dizziness syndromes [12]. In addition, structural connectome studies have demonstrated that the vestibular cortical network is widely distributed and includes multimodal sensory and motor-related regions [40]. The decreased cerebellar and temporal，central connectivity observed here may reflect reduced coupling between cerebellar and sensorimotor systems. However, his finding should be interpreted cautiously, gamma-band EEG is susceptible to muscle and other non-neural artifacts, and scalp-level EEG has limited ability to localize cerebellar generators.

The main contribution of this study is that regional spectral power, wPLI-based connectivity, posturography, and psychological measures were examined within the same cohort. This multimodal design provides preliminary evidence that resting-state EEG alterations in PPPD are related to clinically relevant features, including postural instability and anxiety symptoms. These findings indicate that PPPD is associated with selective, frequency-dependent reorganization of functional brain networks involving posterior sensory regions, frontal control systems, and cerebellar–sensorimotor circuits. This pattern aligns with contemporary models of PPPD emphasizing abnormal sensory reweighting and predictive processing dysfunction across distributed cortical and subcortical networks [5,6].

Several limitations should be acknowledged. First, the sample size was modest, with 40 patients and 40 controls, which may limit statistical power and the generalizability of the findings. Second, this was a single-center study, and potential center-specific recruitment or assessment bias cannot be excluded. Third, the cross-sectional design precluded causal inference regarding whether the observed EEG abnormalities represent predisposing factors, compensatory changes, or consequences of persistent dizziness. Fourth, no intervention or longitudinal follow-up was included; therefore, the sensitivity of the identified EEG markers to treatment response remains unknown. Fifth, EEG was recorded only in the eye-closed resting state, and task-based EEG during visual motion, vestibular stimulation, or postural challenge was not performed. Sixth, scalp-level EEG has limited spatial resolution, and the present findings should not be interpreted as precise localization of deep or cerebellar generators without source-level validation.

5. Conclusions

Patients with PPPD showed increased theta-band power over frontal and parieto-occipital regions, frequency-specific alterations in functional connectivity, impaired postural stability, and an association between frontal electrophysiological activity and anxiety severity. These findings suggest that PPPD may involve abnormal resting-state interactions among cognitive-affective, visual, vestibular-related, and sensorimotor systems. Future studies should use larger multicenter cohorts, longitudinal designs, source-localized EEG or multimodal EEG–fMRI approaches, and task paradigms involving visual motion or vestibular stimulation. Interventional studies, including vestibular rehabilitation, cognitive-behavioral therapy, neuromodulation, or combined treatment strategies, are also needed to determine whether these electrophysiological alterations are modifiable and whether they can serve as biomarkers for prognosis or treatment response.