Cerebral responses to in-game failure and resting-state connectivity are jointly associated with gaming addiction

Cerebral responses to in-game failure and resting-state connectivity are jointly associated with gaming addiction | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Cerebral responses to in-game failure and resting-state connectivity are jointly associated with gaming addiction Jonathan M. Dong, Feng-Rui Zhang This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7044193/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 7 You are reading this latest preprint version Abstract Internet Gaming Disorder (IGD) is a growing behavioral addiction among adolescents and young adults, yet its neurophysiological underpinnings remain unclear. This study aimed to identify EEG biomarkers of IGD by integrating event-related potentials (ERPs) during naturalistic gameplay with resting-state brain connectivity. We analyzed EEG responses to in-game defeat and slay events in a multiplayer online battle arena (MOBA) game and computed connectivity metrics (i.e., weighted phase lag index, clustering coefficient, path length, and global efficiency) across frequency bands during eyes-closed and eyes-open resting-state conditions. Correlation analyses revealed that neural responses to defeat, particularly beta-band synchronization and ERP amplitudes, were significantly associated with IGD severity. Additionally, delta-band connectivity during eyes-closed rest showed robust associations with multiple IGD subscales. Canonical correlation analysis further demonstrated a significant multivariate relationship between a combined neural signature and IGD severity. These findings suggest that both reactive neural responses to negative in-game events and intrinsic functional connectivity jointly contribute to individual vulnerability to gaming addiction. This multidimensional EEG framework offers novel insights into the neurobiological mechanisms of IGD and holds promise for improving assessment and intervention strategies. Biological sciences/Neuroscience Biological sciences/Psychology Social science/Psychology gaming addiction MOBA game EEG resting-state ERP Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Introduction In recent years, electronic sports (esports) have gained tremendous popularity among young people. Over the past decade, certain competitive genres, particularly multiplayer online battle arena (MOBA) games, have not only attracted global attention but have also been formally recognized in international sporting events 1 , 2 . While esports have been shown to enhance attentional control, motor precision, and emotion regulation 1 , they also carry significant psychological risks, most notably, the development of game addiction 3 , 4 . This is particularly concerning among adolescents and young adults, where the prevalence of Internet Gaming Disorder (IGD) is notably high 5 . IGD is considered a behavioral addiction, characterized by compulsive and uncontrollable online gaming, which can severely impair academic performance, social functioning, and physical and mental health 6 . Increasing evidence has linked IGD to functional and connectivity alterations in the brain 7 . Therefore, identifying neural markers specific to IGD is critical for improving clinical diagnosis and intervention strategies. Although previous studies on gaming addiction have revealed several neural correlates, such as altered activity in the anterior cingulate 7 , changes in spectral power 8 , these findings predominantly stem from resting-state data and often rely on broad, non-specific measures. A major gap remains in understanding how real-time neural responses during gameplay contribute to the manifestation of IGD. To address this, an integrative approach combining event-related and resting-state EEG analyses is warranted to capture both transient neural dynamics and stable brain connectivity features. Event-related potentials (ERPs), derived from electroencephalography (EEG) signals time-locked to specific in-game events, provide insights into the brain’s rapid evaluation of motivationally salient outcomes such as victory or defeat. In parallel, resting-state EEG reflects intrinsic, trait-like neural activity and organization. Brain network connectivity metrics, including weighted phase lag index (WPLI), clustering coefficient, characteristic path length, and global efficiency, further enable the assessment of functional integration across frequency bands during resting-state 9 . Together, these EEG-derived signatures form a multidimensional framework for characterizing both state-dependent and trait-like neural mechanisms underlying gaming addiction. In the present study, we aim to identify EEG biomarkers associated with IGD by jointly analyzing event-related EEG responses to in-game events and resting-state brain connectivity across frequency bands. we analyzed ERP responses related to in-game slay and defeat events, including both time-domain and time–frequency components, as well as brain connectivity features (WPLI, clustering coefficient, path length, and global efficiency) from eyes-closed and eyes-open resting-state EEG. We examined their correlations with IGD-20 scores and subscales, and further applied canonical correlation analysis (CCA) to quantify the relationship between ERP and resting-state features and IGD severity. We identified defeated-ERPs and eyes-closed delta-band connectivity features are correlated with IGD. These findings suggest that both stable brain network organization and context-specific neural responses jointly shape individual vulnerability to gaming addiction. Results Self-reported gaming behavior and IGD-20 scores To assess gaming addiction, both subjective and objective indicators were evaluated. The subjective measure consisted of self-reported weekly gaming hours for Honor of Kings , which varied widely across subjects (Fig. 1 A). As an objective measure, the IGD-20 was used to assess six dimensions of gaming disorder: salience, mood modification, tolerance, withdrawal, conflict, and relapse. Total IGD-20 scores showed substantial variability across subjects (Fig. 1 B), with distinct distributions observed across the six subscales (Fig. 1 C to H). To examine the relationship between subjective gaming behavior and gaming disorder severity, we conducted correlation analyses between weekly gaming hours and IGD-20 total scores. The results revealed no significant association between these two measures (Table S1 ), suggesting that self-reported gaming duration may not reliably reflect the severity of gaming addiction. Defeated-induced cerebral responses are associated with gaming addiction To investigate real-time neural responses during gameplay, we analyzed ERPs and time-frequency dynamics in response to two types of in-game events: slay (the subject slayed an opponent) and defeated (the subject was defeated by an opponent). In the time domain, both event types elicited a positive ERP component peaking between 300 and 400 ms post-event, with topographic maps indicating maximal activity over frontal-parietal regions (Fig. 2 A and S1). Time-frequency analysis revealed alpha-band event-related desynchronization between 600 and 1000 ms post-event for both event types, and a clear beta-band event-related synchronization only following the defeated events between 300 and 1000 ms (Fig. 2 B). Statistical comparisons showed that the ERP amplitude in the defeated condition was significantly larger than in the slay condition (p < 0.0001; Fig. 2 C). Beta-band power was significantly greater following defeated events than slay events (p < 0.0001; Fig. 2 D). However, no significant difference was observed in alpha-band power between the two conditions (p = 0.4368; Fig. 2 E). We next examined the correlations between IGD-20 scores and all EEG signatures (i.e., ERP amplitudes, beta power, and alpha power) for both slay and defeated events (Table S2). Pearson correlation analyses revealed that only defeated-related neural responses were significantly associated with IGD-20 scores (Fig. 3 ). Specifically, alpha-band power was positively correlated with the IGD total score (r = 0.4609, p = 0.0269; Fig. 3 A), conflict score (r = 0.5028, p = 0.0145; Fig. 3 B), and salience score (r = 0.4597, p = 0.0273; Fig. 3 C). In contrast, ERP amplitude was negatively correlated with the relapse subscale (r = -0.4555, p = 0.0290; Fig. 3 D), and beta-band power was positively correlated with the conflict subscale (r = 0.4520, p = 0.0303; Fig. 3 E). These results suggest that neural responses to defeated events, rather than slay events, may reflect gaming addiction. Resting-state brain connectivity features are related to gaming addiction Resting-state EEG has been widely employed to characterize neurophysiological signatures of gaming disorder 10 . While previous studies have primarily focused on spectra 8 , micro state 11 , and coherence analysis 12 , we extended this analysis by examining functional connectivity using the WPLI and derived graph-theoretical metrics, including clustering coefficient, average path length, and global efficiency. Figure 4 presents resting-state EEG brain connectivity features under both eyes-open and eyes-closed conditions across five frequency bands. Notably, in the alpha and beta bands, the eyes-closed condition showed significantly higher whole-brain WPLI (Fig. 4 A; Table S3A), clustering coefficient (Fig. 4 B; Table S3B), and global efficiency (Fig. 4 D; Table S3D), along with significantly lower mean path length compared to the eyes-open condition (Fig. 4 C; Table S3C). These results suggest that brain networks are more integrated and efficient during the eyes-closed resting state. We next examined associations between IGD-20 scores and resting-state brain connectivity features (Table S4). Notably, alpha-band connectivity in the eyes-open condition was consistently correlated with the conflict subscale of the IGD-20, suggesting that alpha-band network features may serve as reliable neural markers of gaming-related interpersonal conflict (Fig. 5 ). In contrast, delta-band connectivity in the eyes-closed condition showed widespread associations with IGD severity. Specifically, WPLI was positively correlated with the IGD total score, tolerance, withdrawal symptoms, and relapse (Fig. 5 A); clustering coefficient was positively correlated with the total score, tolerance, and relapse (Fig. 5 B); mean path length was negatively correlated with the total score, mood modification, and relapse (Fig. 5 C); and global efficiency was positively correlated with the total score, withdrawal symptoms, and relapse (Fig. 5 D). These findings suggest that low-frequency functional integration during rest is associated with multiple behavioral dimensions of gaming addiction. In addition, several scattered correlations were observed, including an association between gaming hours and delta-band clustering coefficient in the eyes-open condition (Fig. 5 B), between IGD total score and relapse with theta-band mean path length in the eyes-closed condition (Fig. 5 C), and between conflict score and beta-band global efficiency in the eyes-open condition (Fig. 5 D). Collectively, these results highlight the importance of resting-state connectivity, particularly in the delta frequency bands, as potential electrophysiological markers of gaming addiction. Canonical correlation between neural features and gaming disorder To examine the multivariate relationship between EEG features and IGD severity, we conducted a CCA (Fig. 6 ). Based on prior findings, we selected delta-band brain connectivity features from the eyes-closed resting-state condition, i.e., WPLI, clustering coefficient, mean path length, and global efficiency, as representative measures of intrinsic functional integration, given their consistent associations with IGD scores. In parallel, we selected defeated-related ERP features, including amplitude, beta-band power, and alpha-band power within their ROIs, as representative task-evoked neural responses. To reduce dimensionality, principal component analysis (PCA) was performed separately on the resting-state and ERP features. The first principal component (PC1) from each set was extracted, yielding two orthogonal neural features. These two neural features were then entered into the CCA model to examine their joint relationship with IGD severity, as measured by the total IGD-20 score. CCA revealed a significant canonical correlation between neural features and IGD score (r = 0.662, p = 0.0031, Wilks’ Lambda), indicating a strong multivariate association. Canonical loadings indicated that delta-band brain connectivity contributed more strongly to the canonical variate (loading = 0.876), while the defeated-ERP component contributed moderately (loading = -0.522). A bootstrap analysis with 1,000 iterations confirmed the robustness of this relationship, yielding a 95% confidence interval ranging from 0.441 to 0.863. These results suggest that IGD severity is reliably associated with a combined neural signature encompassing both resting-state connectivity and event-related EEG responses. Discussion The present study integrated psychometric assessments, task-related EEG, and resting-state brain connectivity analyses to investigate the neurophysiological underpinnings of gaming addiction in adolescents and young adults. By leveraging a naturalistic MOBA gameplay paradigm and a publicly available EEG dataset, we explored both evoked neural responses to in-game events and intrinsic functional network properties during rest. Our findings revealed that electrophysiological features, particularly those elicited during defeat events and resting-state delta-band connectivity, were significantly associated with the severity of IGD-20. CCA further demonstrated a robust multivariate relationship between these neural markers and IGD scores, suggesting that gaming addiction is reflected in a combined neural signature encompassing both reactive and intrinsic brain processes. Subjective self-reported gaming hour may have bias An eccentric observation from our analysis is the lack of correlation between self-reported weekly gaming hours and IGD-20 total scores. The gaming hours are often used as a proxy for gaming addiction 13 , while our findings suggest that this measure may not accurately capture the psychological dimensions of gaming addiction. One possible explanation is that self-reported gaming time was subject to bias, particularly social desirability bias 14 , 15 . During participation in formal experiments, especially those involving neural recordings or addiction-related assessments, participants may intentionally underreport their gaming duration in an effort to present themselves in a more favorable light. Additionally, retrospective recall inaccuracies may further compromise the reliability of self-reported time estimates. In contrast, standardized psychometric instruments like the IGD-20 capture the broader emotional, cognitive, and behavioral features of gaming disorder, providing a more stable and comprehensive assessment 16 . This discrepancy underscores the importance of incorporating validated diagnostic scales rather than relying solely on subjective behavioral metrics when evaluating gaming-related pathology. Defeated-induced non-phase locking oscillations are sensitive to gaming disorder Our task-based EEG analyses revealed that defeat-related neural responses, rather than those associated with slay, were more strongly correlated with IGD severity. In particular, alpha- and beta-band power were positively correlated with gaming disorder severity, and ERP amplitude was negatively related to relapse score. The non-phase locking oscillations, especially the beta-ROI, possible is related to cognitive and emotional processing 17 , suggesting the stronger emotional responses to negative event predicts a serious gaming disorder. These findings align with prior studies suggesting that individuals with gaming addiction exhibit heightened emotional reactivity to negative outcomes, worth emotional regulation, and impaired cognitive control in response to in-game stressors 18 – 21 . Notably, these event-related EEG markers may provide an ecologically valid window into how individuals with IGD process failure and loss during competitive gameplay. Brain connectivity features potentially serve as the biomarkers of gaming addiction Alpha-band connectivity features during the eyes-open resting state (including whole-brain WPLI, clustering coefficient, and global efficiency) were significantly associated with the conflict subscale of the IGD-20. This subscale measures the extent to which gaming behavior disrupts interpersonal relationships, academic performance, or occupational functioning. Conceptually, it reflects impairments in self-regulation, social-emotional monitoring, and executive control in real-world contexts 22 – 24 . alpha-band network activity is associated with cognitive control, information integration, conflict monitoring, and emotional regulation 25 , 26 . Therefore, the observed negative associations suggest that individuals with higher levels of gaming-related conflict exhibit weakened alpha-band network integration during eyes-open rest. This finding supports the interpretation that reduced alpha-band connectivity may reflect impairments in externally directed attention and social-emotional regulation, particularly in contexts involving interpersonal or functional conflict. In contrast, delta-band connectivity metrics during the eyes-closed resting state showed consistent with multiple behavioral dimensions of IGD. Delta-band components are usually associated with large-scale neural integration 27 , and are often implicated in neuropsychiatric conditions such as addiction and depression 27 , 28 . In this study, individuals with higher IGD severity exhibited greater delta-band global efficiency and clustering coefficient, along with shorter average path length, suggesting a more compact and integrated network configuration. This compensatory integration may reflect a neuroadaptive reorganization of brain systems involved in motivation, emotional regulation, and cognitive control as a result of extensive gaming exposure. Given that eyes-closed resting-state EEG captures more intrinsic, trait-like brain activity 29 , delta-band functional connectivity may serve as a reliable and stable neurophysiological biomarker of gaming addiction. Taken together, the distinct yet complementary patterns observed in alpha and delta bands across different resting-state conditions suggest frequency-specific and state-dependent neural correlates of IGD. While alpha-band alterations may relate more to attention and social conflict regulation in externally engaged states, delta-band connectivity reflects broader intrinsic changes in large-scale network integration. Future studies should test the predictive validity of these band-specific features in larger and longitudinal cohorts and explore their responsiveness to interventions such as cognitive control training or neurofeedback therapy. Selection and dimensionality reduction of neural features for CCA To examine the joint neural correlates of gaming addiction, we conducted a CCA using electrophysiological features drawn from two domains: event-related EEG responses and resting-state connectivity. Specifically, we selected defeated-related ERP features and delta-band resting-state connectivity for inclusion in the model. This operation was based on prior univariate analyses showing that these two neural domains demonstrated the strongest and most consistent associations with IGD severity. Defeated-related EEG features were significantly correlated with various IGD-20 subscales, particularly relapse and conflict. Likewise, delta-band connectivity metrics from the eyes-closed resting state showed robust correlations with multiple IGD dimensions. These findings collectively indicated that both reactive neural responses to in-game failure and intrinsic low-frequency network organization are tightly linked to the behavioral manifestations of gaming addiction. Given the relatively small sample size (n = 23) and the high dimensionality of the input features, we applied PCA to each neural domain to reduce the number of variables and improve the robustness of the CCA estimation. From each domain, only the first principal component was retained, capturing the dominant variance within defeated-related ERP features and delta-band connectivity features, respectively. This dimensionality reduction strategy minimizes the risk of overfitting and ensures stable estimation of canonical variates in small-sample multivariate models 30 , 31 . The resulting two neural components were then used to represent the neural dataset in the CCA. Limitation This study has two primary limitations that should be considered when interpreting the findings. First, although the sample size (n = 23) yielded acceptable statistical power (approximately 0.71 under the assumptions of r = 0.5 and α = 0.05), it remains modest and limits the generalizability of the results. The small sample size also constrained our ability to apply more advanced analytical methods, such as structural equation modeling or machine learning-based classification approaches, which typically require larger datasets to ensure model stability and prevent overfitting 32 , 33 . Future studies with larger and more diverse samples would allow for more comprehensive modeling of the complex relationships between neural dynamics, psychological traits, and gaming behavior. Second, the cross-sectional design of this study prevents causal inference regarding the directionality of the observed brain–behavior associations. The identified defeated-related ERP responses and delta-band connectivity features may either reflect predisposing neural risk markers for Internet Gaming Disorder or emerge as neuroplastic adaptations resulting from prolonged gaming. Longitudinal or intervention-based designs that monitor electrophysiological changes over time are necessary to disentangle these possibilities and to determine the predictive utility of the identified neural features. Third, the dataset used in this study was drawn exclusively from Chinese young adults, a population with a distinct cultural context regarding gaming norms, social expectations, and education systems 34 . Cultural factors are known to influence both the prevalence and expression of gaming disorder 35 , as well as the neural and behavioral responses associated with gaming experiences. Therefore, caution is warranted in generalizing these findings to other populations. Future studies should aim to replicate and extend the current results across more diverse cultural and demographic groups to examine the cross-cultural robustness and universality of the identified EEG markers. Conclusion The present study provides novel evidence linking both task-evoked and resting-state electrophysiological features to the severity of IGD-20 in young individuals. We demonstrated that neural responses to defeat during naturalistic gameplay, specifically ERP amplitude and post-event beta and alpha power—were significantly associated with key dimensions of gaming addiction, including relapse and conflict. Additionally, resting-state connectivity in the delta frequency band, particularly under eyes-closed conditions, was strongly correlated with multiple IGD subscales, suggesting its potential as a stable neural marker of intrinsic functional organization in addiction. Multivariate analysis via canonical correlation further revealed that a joint neural signature combining defeated-related ERP features and delta-band connectivity robustly predicted IGD severity. Together, these findings highlight the utility of integrating time-locked and intrinsic EEG features to capture the multifaceted neural underpinnings of gaming addiction and offer promising targets for future diagnostic and intervention efforts. Methods Participants and procedure 23 healthy subjects (8 female and 15 male participants), aged from 15 to 22 years, were recruited for this study. Each subject reported their average weekly gaming hours and completed the Internet Gaming Disorder Scale-20 (IGD-20). Both the total score and six subscale scores (salience, mood modification, tolerance, withdrawal, conflict, and relapse) were recorded for further analysis. Subjects were instructed to play 5 to 7 rounds of the multiplayer online battle arena (MOBA) game Honor of Kings under naturalistic conditions 36 . EEG recording Electroencephalography (EEG) data were obtained from a publicly available dataset published by Li et al. on OpenNeuro 36 . EEG signals were recorded using a 64-channel Neuroscan system at a sampling rate of 1000 Hz. Two types of EEG data were acquired: (1) gameplay recordings with event markers for game start, game end, slaying an enemy, and being defeated by an enemy; and (2) resting-state recordings (eyes-open and eyes-closed) lasting approximately 6 to 7 minutes. EEG preprocessing EEG preprocessing was conducted using EEGLAB in MATLAB 37 . Both game-playing and resting-states EEG signals were band-pass filtered between 1 and 45 Hz and re-referenced to the average of the M1 and M2 electrodes. Noisy channels, identified as having power spectral values more than three standard deviations from the mean, were removed and interpolated using spherical spline interpolation. Independent component analysis (ICA) was used to remove ocular and muscle artifacts, following procedures 38 . For the gameplay data, epochs were extracted from − 500 ms to 1500 ms relative to two event markers: "slay" (when the participant slayed an enemy) and "defeated" (when the participant was defeated by an enemy). Baseline correction was performed using the pre-event window (-500 to 0 ms). For resting-state EEG, a 200-second segment from the middle of the recording was selected for subsequent analysis. Event-related EEG components extraction Event-related potentials (ERPs) were obtained by averaging the epoched data filtered from 1 to 30 Hz. ERPs were computed over a region of interest (ROI) comprising nine frontal-parietal electrodes (F1, Fz, F2, FC1, FCz, FC2, C1, Cz, and C2). For each condition, the ERP peak amplitudes were extracted as the maximum value occurring between 200 and 400 ms post-stimulus. Time-frequency distributions (TFDs) were computed using MATLAB’s built-in function spectrogram , with a 200-ms Hanning window, 190 ms overlap of widow, and 1000 nfft, yielding a time resolution of 10 ms and frequency resolution of 1 Hz. Baseline correction was applied to each trial by normalizing the TFDs using the − 200 to 0 ms pre-stimulus interval. Beta-band power (14 to 30 Hz) was averaged over the ROI during 300 to 1000 ms post-event, and alpha-band power (8 to 13 Hz) was averaged over 600 to 1000 ms. Resting-state EEG brain connectivity features extraction Resting-state EEG data were analyzed using the Brain Connectivity Toolbox in MATLAB 39 . Signals were filtered into five frequency bands: delta (1 to 4 Hz), theta (4 to 8 Hz), alpha (8 to 13 Hz), beta (13 to 30 Hz), and gamma (31 to 45 Hz). For each band, the weighted phase lag index (WPLI) was computed between all pairs of 60 electrodes to estimate phase synchronization while minimizing volume conduction 38 . The mean of all pairwise wPLI values was used to represent whole-brain functional connectivity 40 . To characterize the topological properties of the brain network, three graph-theoretical metrics were derived from the wPLI matrices: clustering coefficient 41 , average path length 42 , and global efficiency 43 . Distance matrices were calculated as the inverse of the WPLI matrices (with appropriate handling of zeros and infinities), and used to compute shortest paths. The clustering coefficient was averaged across nodes, while average path length and global efficiency were computed from the resulting distance matrices. Canonical correlation analysis To examine the relationship between gaming disorder and neural indices, a canonical correlation analysis (CCA) was conducted in R using the CCP package. The gaming disorder consisted of the IGD total score, while the neural variable set included electrophysiological measures derived from principal component analysis (PCA) of defeated-related components (ERP amplitude, beta ROI power, alpha ROI power) and delta-band brain connectivity features (whole brain WPLI, clustering coefficient, average path length, and global efficiency). Specifically, PCA was first applied separately to defeated-related components and delta-band brain connectivity features. The first principal component (PC1) of defeated ERP features (ERP_PC1) and the first PC of delta band brain connectivity features (delta_PC1) were extracted and used as summary indices. These two components represented the neural data matrix Y , while the IGD total score served as the behavioral matrix X . Both X and Y were standardized prior to analysis. CCA was then performed to identify the maximal linear relationship between brain activities and gaming disorder. Canonical loadings, defined as the Pearson correlation coefficients between the original neural variables (ERP_PC1 and delta_PC1) and the corresponding canonical variate, were also calculated to assess variable contributions. Statistical analyses All statistical analyses were conducted using GraphPad Prism (version 9) and jamovi (version 2.3). Paired-sample t tests were used to compare ERP amplitude, alpha ROI power, and beta ROI power between the slay and defeated conditions. Brain connectivity features were analyzed using two-way analysis of variance (ANOVA) with Bonferroni post hoc correction. Pearson correlation and linear regression analyses were performed to examine relationships between IGD-20 scores, gameplay behavior, and EEG-derived features. For the CCA results, statistical significance of the canonical correlation coefficient was evaluated using Wilks’ lambda approximation. To assess the robustness of the first canonical correlation, a non-parametric bootstrap analysis with 1000 iterations was conducted by resampling with replacement. The resulting bootstrap distribution was used to compute the 95% confidence interval (CI), defined by the 2.5th and 97.5th percentiles. Declarations Ethics Statement This study was approved by the Ethics Committee of University-Town Hospital of Chongqing Medical University (Approval No. LL-202307). The EEG data analyzed in this manuscript were obtained from a previously published, open-access dataset (Li et al., 2024, OpenNeuro accession number: ds005520; Li et al., 2025, Scientific Data). All procedures involving human participants in the original study were conducted in accordance with the ethical standards of the institutional research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. Participant Consent In the original study from which the EEG data used in this manuscript were obtained (Li et al., 2024, OpenNeuro accession number: ds005520; Li et al., 2025, Scientific Data), all participants were fully informed about the study procedures, potential risks, and their rights prior to enrollment. Written informed consent to participate in the study and for the subsequent use and publication of their anonymized data was obtained from each participant under the oversight of the Ethics Committee of University-Town Hospital of Chongqing Medical University (Approval No. LL-202307). For the current secondary analysis, no new human data were collected, and the authors only accessed and analyzed de-identified, open-access data. Acknowledgements We thank Hong-Zhi Li, Jia-Jia Yang, Zhen Lv, Li-Yang Wan, Wo Wang, Da-Qi Li, Dong-Dong Zhou, ang Li Kuang for the data support. Funding declaration: The authors received no specific funding for this work. Author contributions: Feng-Rui Zhang designed study. Jonathan M. Dong and Feng-Rui Zhang analyzed the data and wrote the manuscript. Conflicts of interest : None. Data and script availability statement : The dataset used in this study is publicly available on OpenNeuro (https://openneuro.org/datasets/ds005520/versions/1.0.0). All analysis scripts are available from the corresponding author upon reasonable request. References Mora-Cantallops, M. & Sicilia, M.-Á. MOBA games: A literature review. Entertainment Computing 26 , 128-138, doi: 10.1016/j.entcom.2018.02.005 (2018). Cui, L., Kim, E. J. & Kim, J. How Chinese Media Addresses Esports Issues: A Text Mining Comparative Analysis of Online News and Viewers’ Comments on the Hangzhou Asian Games. Electronics 12 , 4961 (2023). Pontes, H. M., Rumpf, H.-J., Selak, Š. & Montag, C. Investigating the interplay between gaming disorder and functional impairments in professional esports gaming. Sci Rep-Uk 14 , 6557, doi:10.1038/s41598-024-56358-x (2024). Wischert-Zielke, M. & Barke, A. Differences between recreational gamers and Internet Gaming Disorder candidates in a sample of Animal Crossing: New Horizons players. Sci Rep-Uk 13 , 5102, doi:10.1038/s41598-023-32113-6 (2023). Gao, Y.-X., Wang, J.-Y. & Dong, G.-H. The prevalence and possible risk factors of internet gaming disorder among adolescents and young adults: Systematic reviews and meta-analyses. Journal of Psychiatric Research 154 , 35-43, doi: 10.1016/j.jpsychires.2022.06.049 (2022). Petry, N. M. et al. An international consensus for assessing internet gaming disorder using the new DSM-5 approach. Addiction 109 , 1399-1406, doi: 10.1111/add.12457 (2014). Zhang, J.-T. et al. Altered resting-state functional connectivity of the insula in young adults with Internet gaming disorder. Addiction Biology 21 , 743-751, doi: 10.1111/adb.12247 (2016). Park, S., Ha, J., Ahn, W. & Kim, L. Measurement of craving among gamers with internet gaming disorder using repeated presentations of game videos: a resting-state electroencephalography study. BMC Public Health 23 , 816, doi:10.1186/s12889-023-15750-4 (2023). Wang, Z. et al. Cortical complexity and connectivity during isoflurane-induced general anesthesia: a rat study. Journal of Neural Engineering 19 , 036009, doi:10.1088/1741-2552/ac6a7b (2022). Burleigh, T. L., Griffiths, M. D., Sumich, A., Wang, G. Y. & Kuss, D. J. Gaming disorder and internet addiction: A systematic review of resting-state EEG studies. Addictive Behaviors 107 , 106429, doi: 10.1016/j.addbeh.2020.106429 (2020). Kim, J. S. et al. Resting-state EEG microstate analysis of internet gaming disorder and alcohol use disorder. Dialogues in Clinical Neuroscience 26 , 89-102, doi:10.1080/19585969.2024.2432913 (2024). Lee, J.-Y., Choi, J.-S. & Kwon, J. S. Neurophysiological Mechanisms of Resilience as a Protective Factor in Patients with Internet Gaming Disorder: A Resting-State EEG Coherence Study. Journal of Clinical Medicine 8 , 49 (2019). Fuster, H., Carbonell, X., Pontes, H. M. & Griffiths, M. D. Spanish validation of the Internet Gaming Disorder-20 (IGD-20) Test. Computers in Human Behavior 56 , 215-224, doi: 10.1016/j.chb.2015.11.050 (2016). Krumpal, I. Determinants of social desirability bias in sensitive surveys: a literature review. Quality & Quantity 47 , 2025-2047, doi:10.1007/s11135-011-9640-9 (2013). King, M. F. & Bruner, G. C. Social desirability bias: A neglected aspect of validity testing. Psychology & Marketing 17 , 79-103, doi: 10.1002/(SICI)1520-6793(200002)17:23.0.CO;2-0 (2000). Shu M., Y., Ivan Jacob, A. P., Meng Xuan, Z. & Anise M. S., W. Psychometric validation of the Internet Gaming Disorder-20 Test among Chinese middle school and university students. Journal of Behavioral Addictions 8 , 295-305, doi: 10.1556/2006.8.2019.18 (2019). Güntekin, B. & Başar, E. Event-related beta oscillations are affected by emotional eliciting stimuli. Neuroscience Letters 483 , 173-178, doi: 10.1016/j.neulet.2010.08.002 (2010). Yen, J.-Y. et al. Emotional Regulation in Young Adults with Internet Gaming Disorder. International Journal of Environmental Research and Public Health 15 , 30 (2018). Estupiñá, F. J., Bernaldo-de-Quirós, M., Vallejo-Achón, M., Fernández-Arias, I. & Labrador, F. Emotional regulation in Gaming Disorder: A systematic review. The American Journal on Addictions 33 , 605-620, doi: 10.1111/ajad.13621 (2024). Dong, G., Li, H., Wang, L. & Potenza, M. N. Cognitive control and reward/loss processing in Internet gaming disorder: Results from a comparison with recreational Internet game-users. European Psychiatry 44 , 30-38, doi:10.1016/j.eurpsy.2017.03.004 (2017). Wang, Z. et al. Diversiform Etiologies for Post-stroke Depression. Frontiers in Psychiatry Volume 9 - 2018 , doi:10.3389/fpsyt.2018.00761 (2019). Griffiths, M. D. A ‘components’ model of addiction within a biopsychosocial framework. Journal of Substance Use 10 , 191-197, doi: 10.1080/14659890500114359 (2005). Faraci, P., Bottaro, R., Nasonte, G., Valenti, G. D. & Craparo, G. A New Short Version of Internet Gaming Disorder-20: An Exploratory Structural Equation Modeling. Clin Neuropsychiatry 20 , 205-218, doi:10.36131/cnfioritieditore20230305 (2023). Sprong, M. E., Griffiths, M. D., Lloyd, D. P., Paul, E. & Buono, F. D. Comparison of the Video Game Functional Assessment-Revised (VGFA-R) and Internet Gaming Disorder Test (IGD-20). Frontiers in Psychology Volume 10 - 2019 , doi:10.3389/fpsyg.2019.00310 (2019). Wu, S. et al. Disrupted Functional Brain Network Architecture in Sufferers with Boxing-Related Repeated Mild Traumatic Brain Injury: A Resting-State EEG Study. JIN 23 , doi:10.31083/j.jin2305102 (2024). Lejko, N., Larabi, D. I., Herrmann, C. S., Aleman, A. & Ćurčić-Blake, B. Alpha Power and Functional Connectivity in Cognitive Decline: A Systematic Review and Meta-Analysis. Journal of Alzheimer’s Disease 78 , 1047-1088, doi:10.3233/jad-200962 (2020). Schwartzmann, B. et al. Resting-state EEG delta and alpha power predict response to cognitive behavioral therapy in depression: a Canadian biomarker integration network for depression study. Sci Rep-Uk 13 , 8418, doi:10.1038/s41598-023-35179-4 (2023). Newson, J. J. & Thiagarajan, T. C. EEG Frequency Bands in Psychiatric Disorders: A Review of Resting State Studies. Frontiers in Human Neuroscience Volume 12 - 2018 , doi:10.3389/fnhum.2018.00521 (2019). Costumero, V., Bueichekú, E., Adrián-Ventura, J. & Ávila, C. Opening or closing eyes at rest modulates the functional connectivity of V1 with default and salience networks. Sci Rep-Uk 10 , 9137, doi:10.1038/s41598-020-66100-y (2020). Singanamalli, A. et al. Cascaded Multi-view Canonical Correlation (CaMCCo) for Early Diagnosis of Alzheimer’s Disease via Fusion of Clinical, Imaging and Omic Features. Sci Rep-Uk 7 , 8137, doi:10.1038/s41598-017-03925-0 (2017). Helmer, M. et al. On the stability of canonical correlation analysis and partial least squares with application to brain-behavior associations. Communications Biology 7 , 217, doi:10.1038/s42003-024-05869-4 (2024). Wolf, E. J., Harrington, K. M., Clark, S. L. & Miller, M. W. Sample Size Requirements for Structural Equation Models:An Evaluation of Power, Bias, and Solution Propriety. Educational and Psychological Measurement 73 , 913-934, doi:10.1177/0013164413495237 (2013). Li, J., Horiguchi, Y. & Sawaragi, T. in 2020 59th Annual Conference of the Society of Instrument and Control Engineers of Japan (SICE). 826-831. Liao, Z. et al. Prevalence of Internet Gaming Disorder and Its Association With Personality Traits and Gaming Characteristics Among Chinese Adolescent Gamers. Frontiers in Psychiatry Volume 11 - 2020 , doi:10.3389/fpsyt.2020.598585 (2020). Chew, Peter K. H., Lin, Patrick K. F. & Yow, Yong J. Cross-Cultural Differences in the Pathways to Internet Gaming Disorder. Asia-Pacific Psychiatry 16 , e12565, doi: 10.1111/appy.12565 (2024). Li, H.-Z. et al. (OpenNeuro, 2024). Delorme, A. & Makeig, S. EEGLAB: an open source toolbox for analysis of single-trial EEG dynamics including independent component analysis. Journal of neuroscience methods 134 , 9-21, doi:10.1016/j.jneumeth.2003.10.009 (2004). Zhang, F. et al. Cross-Species Investigation on Resting State Electroencephalogram. Brain Topography 32 , 808-824, doi:10.1007/s10548-019-00723-x (2019). Rubinov, M. & Sporns, O. Complex network measures of brain connectivity: Uses and interpretations. NeuroImage 52 , 1059-1069, doi: 10.1016/j.neuroimage.2009.10.003 (2010). Imperatori, L. S. et al. Cross-participant prediction of vigilance stages through the combined use of wPLI and wSMI EEG functional connectivity metrics. Sleep 44 , doi:10.1093/sleep/zsaa247 (2020). Omurtag, A. et al. Disruption of functional network development in children with prenatal Zika virus exposure revealed by resting-state EEG. Sci Rep-Uk 15 , 6346, doi:10.1038/s41598-025-90860-0 (2025). Mehraram, R., Kries, J., De Clercq, P., Vandermosten, M. & Francart, T. EEG reveals brain network alterations in chronic aphasia during natural speech listening. Sci Rep-Uk 15 , 2441, doi:10.1038/s41598-025-86192-8 (2025). Guo, H. et al. A hybrid critical channels and optimal feature subset selection framework for EEG fatigue recognition. Sci Rep-Uk 15 , 2139, doi:10.1038/s41598-025-86234-1 (2025). Additional Declarations No competing interests reported. Supplementary Files Supplementaryinformation.docx Cite Share Download PDF Status: Under Review Version 1 posted Reviews received at journal 14 Apr, 2026 Reviewers agreed at journal 14 Apr, 2026 Reviewers invited by journal 06 Nov, 2025 Editor assigned by journal 04 Nov, 2025 Editor invited by journal 15 Jul, 2025 Submission checks completed at journal 08 Jul, 2025 First submitted to journal 08 Jul, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7044193","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":504692959,"identity":"6379d8b6-3dee-4a12-8658-95e7a6785574","order_by":0,"name":"Jonathan M. Dong","email":"","orcid":"","institution":"Lower Canada College","correspondingAuthor":false,"prefix":"","firstName":"Jonathan","middleName":"M.","lastName":"Dong","suffix":""},{"id":504692960,"identity":"b34615da-c5fd-4da2-b66f-12d5cb0d732a","order_by":1,"name":"Feng-Rui Zhang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABC0lEQVRIiWNgGAWjYDACCTiLseHAhwobEKPxAJFamBsfzjiTBtFLpBb2ZmPelsNgJl4t8rObnz38wnBP3uAAY5sEb8N5u7Xth4G21NhE49LCOOeYubEMQ7HhBpAWyR23k7edSQRqOZaW24BDC7NEgpm0BEMCI1iL4ZnbyWYHgFoYGw7j1MImkf4NpMUerCWx7Vyy2fmH+LXwSOSYSX5gSEgEamk2ONh2wM7sBgFbJCRyyqQZDBKSZx5mbHzYcCY5wewG0JYEPH6Rn5G+TfJHRYJt3/H2B4f/VNjZm51Pf/jgQ40NTi3gIOAxYGBQgMQIQyJYZQIe5SDA+ANkHdRQewKKR8EoGAWjYAQCACN0ZfdiHXo5AAAAAElFTkSuQmCC","orcid":"","institution":"Washington University School of Medicine","correspondingAuthor":true,"prefix":"","firstName":"Feng-Rui","middleName":"","lastName":"Zhang","suffix":""}],"badges":[],"createdAt":"2025-07-04 07:38:39","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7044193/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7044193/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":91682918,"identity":"e1475649-826b-4ada-b077-0585273a5160","added_by":"auto","created_at":"2025-09-19 06:54:57","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":1744154,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eWeekly gaming duration and Internet Gaming Disorder Scale-20.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(\u003cstrong\u003eA\u003c/strong\u003e) Weekly gaming hours spent on \u003cem\u003eHonor of Kings\u003c/em\u003e. (\u003cstrong\u003eB\u003c/strong\u003e) Total score of Internet Gaming Disorder Scale-20 (IGD-20). (\u003cstrong\u003eC\u003c/strong\u003e to \u003cstrong\u003eH\u003c/strong\u003e) Scores of 6 subscales of IGD-20. For each subfigure, the top panel is a stem plot of individual data, and the bottom panel is a histogram depicting the distribution of data across all 23 participants.\u003c/p\u003e","description":"","filename":"Figure1.png","url":"https://assets-eu.researchsquare.com/files/rs-7044193/v1/f70474f0847707ac4c61cbe3.png"},{"id":91683218,"identity":"8d84b18d-9c67-414f-a738-06a412553d46","added_by":"auto","created_at":"2025-09-19 07:02:57","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":1118166,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eERP components of slay and defeated.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(\u003cstrong\u003eA\u003c/strong\u003e) Time-domain event-related potentials (ERPs) and corresponding scalp topographies evoked by slay and defeated events. (\u003cstrong\u003eB\u003c/strong\u003e) Time-frequency-domain (TFD) spectra for slay- and defeated-related activity. The beta-band region of interest (ROI) spans 14 to 30 Hz from 300 to 1000 ms post-event; the alpha-band ROI spans 8 to 13 Hz from 600 to 1000 ms post-event. (\u003cstrong\u003eC\u003c/strong\u003e to \u003cstrong\u003eE\u003c/strong\u003e) Statistical comparisons (paired-sample \u003cem\u003et \u003c/em\u003etests ) between slay and defeated conditions for ERP amplitudes (C), beta-band power (D), and alpha-band power (E). Data are presented as mean ± SD. Time 0 indicates the onset of slaying an enemy or being defeated by one. ****\u003cem\u003ep\u003c/em\u003e\u0026lt; 0.0001; ns, not significant.\u003c/p\u003e","description":"","filename":"Figure2.png","url":"https://assets-eu.researchsquare.com/files/rs-7044193/v1/42a8abccd545b43f6073f199.png"},{"id":91684004,"identity":"51f56ef9-ffef-41bc-8b7f-91eaa3bd5a1c","added_by":"auto","created_at":"2025-09-19 07:10:57","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":644723,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eCorrelations between ERP components and IGD-20\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(\u003cstrong\u003eA\u003c/strong\u003eto \u003cstrong\u003eC\u003c/strong\u003e) Correlations between defeated-induced alpha ROI power and IGD-20 total score (A), Conflict score (B), and Salience score (C), respectively. (\u003cstrong\u003eD\u003c/strong\u003e) Correlation between defeated-induced ERP amplitude and Relapse score. (\u003cstrong\u003eE\u003c/strong\u003e) Correlation between defeated-induced beta ROI power and Conflict score. Statistics utilized Pearson’s correlation.\u003c/p\u003e","description":"","filename":"Figure3.png","url":"https://assets-eu.researchsquare.com/files/rs-7044193/v1/e7bc2eb1508a5fbe2bb4ad71.png"},{"id":91682922,"identity":"c1cc1d47-5afe-4542-a5ec-6a08bcd495e7","added_by":"auto","created_at":"2025-09-19 06:54:57","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":399018,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eBrain connectivity features from resting-state EEG.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(\u003cstrong\u003eA\u003c/strong\u003eto \u003cstrong\u003eD\u003c/strong\u003e) Whole brain WPLI (A), clustering coefficient (B), mean path length (C), and global efficiency (D) of five frequency bands (delta: 1 to 4 Hz, theta: 4 to 7 Hz, alpha: 8 to 13 Hz, beta: 14 to 30 Hz, gamma: 31 to 45 Hz) for both eyes-closed and eyes-open resting-state EEG. Statistics utilized repeated measures two-way ANOVA, post-hoc used paired-sample \u003cem\u003et\u003c/em\u003e test with Bonferroni correction. *\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.05; **\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.01; ***\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.001; ****\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.0001; ns, not significant.\u003c/p\u003e","description":"","filename":"Figure4.png","url":"https://assets-eu.researchsquare.com/files/rs-7044193/v1/63ed9fe56e3a53299d199a6b.png"},{"id":91684005,"identity":"8ba33a23-44e6-47f6-9561-fd5417bbeec9","added_by":"auto","created_at":"2025-09-19 07:10:57","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":961317,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eCorrelations between brain connectivity features and IGD-20\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(\u003cstrong\u003eA\u003c/strong\u003e to \u003cstrong\u003eD\u003c/strong\u003e) Correlation matrices showing associations between IGD-20 scores and whole-brain connectivity features: WPLI (A), clustering coefficient (B), mean path length (C), and global efficiency (D), across five frequency bands (delta: 1 to 4 Hz, theta: 4 to 7 Hz, alpha: 8 to 13 Hz, beta: 14 to 30 Hz, gamma: 31 to 45 Hz) under both eyes-closed and eyes-open resting-state EEG conditions. Pearson’s correlation was used for all analyses. Black and white circles represent negative and positive correlations, respectively. Statistically significant correlations are highlighted with a red border.\u003c/p\u003e","description":"","filename":"Figure5.png","url":"https://assets-eu.researchsquare.com/files/rs-7044193/v1/24fc3d7f4f1ad18803746d42.png"},{"id":91682919,"identity":"89c0b413-6edf-4f33-9785-7e6471b3f1f0","added_by":"auto","created_at":"2025-09-19 06:54:57","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":217482,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eCanonical correlation analysis between representative brain activities and gaming disorder.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eCanonical correlation analysis (CCA) was conducted to examine the relationship between representative brain activity features and gaming disorder severity. Delta-band brain connectivity of resting-state EEG and ERP features evoked by defeated events were used as neural predictors, while the IGD-20 total score represented gaming disorder. A significant canonical correlation (r = 0.662) between brain features and IGD-20 scores. The canonical loadings were 0.876 for delta-band connectivity and 0.552 for defeated-related ERP components.\u003c/p\u003e","description":"","filename":"Figure6.png","url":"https://assets-eu.researchsquare.com/files/rs-7044193/v1/f96319e698403d48e7615c99.png"},{"id":91817371,"identity":"2f1ecf4b-2493-4440-b083-a05a4423d41d","added_by":"auto","created_at":"2025-09-22 06:55:09","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":6328227,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7044193/v1/056d540e-2375-4bb6-8a9d-43097304e784.pdf"},{"id":91683220,"identity":"62d46ed2-8d9d-4796-a969-5e6dc86cb3b3","added_by":"auto","created_at":"2025-09-19 07:02:57","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":4019313,"visible":true,"origin":"","legend":"","description":"","filename":"Supplementaryinformation.docx","url":"https://assets-eu.researchsquare.com/files/rs-7044193/v1/5f391a98b86820d5cf046bb9.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Cerebral responses to in-game failure and resting-state connectivity are jointly associated with gaming addiction","fulltext":[{"header":"Introduction","content":"\u003cp\u003eIn recent years, electronic sports (esports) have gained tremendous popularity among young people. Over the past decade, certain competitive genres, particularly multiplayer online battle arena (MOBA) games, have not only attracted global attention but have also been formally recognized in international sporting events\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e,\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e. While esports have been shown to enhance attentional control, motor precision, and emotion regulation\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u003c/sup\u003e, they also carry significant psychological risks, most notably, the development of game addiction\u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e,\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u003c/sup\u003e. This is particularly concerning among adolescents and young adults, where the prevalence of Internet Gaming Disorder (IGD) is notably high\u003csup\u003e\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e. IGD is considered a behavioral addiction, characterized by compulsive and uncontrollable online gaming, which can severely impair academic performance, social functioning, and physical and mental health\u003csup\u003e\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u003c/sup\u003e. Increasing evidence has linked IGD to functional and connectivity alterations in the brain\u003csup\u003e\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u003c/sup\u003e. Therefore, identifying neural markers specific to IGD is critical for improving clinical diagnosis and intervention strategies.\u003c/p\u003e\u003cp\u003eAlthough previous studies on gaming addiction have revealed several neural correlates, such as altered activity in the anterior cingulate\u003csup\u003e\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u003c/sup\u003e, changes in spectral power\u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e, these findings predominantly stem from resting-state data and often rely on broad, non-specific measures. A major gap remains in understanding how real-time neural responses during gameplay contribute to the manifestation of IGD. To address this, an integrative approach combining event-related and resting-state EEG analyses is warranted to capture both transient neural dynamics and stable brain connectivity features.\u003c/p\u003e\u003cp\u003eEvent-related potentials (ERPs), derived from electroencephalography (EEG) signals time-locked to specific in-game events, provide insights into the brain\u0026rsquo;s rapid evaluation of motivationally salient outcomes such as victory or defeat. In parallel, resting-state EEG reflects intrinsic, trait-like neural activity and organization. Brain network connectivity metrics, including weighted phase lag index (WPLI), clustering coefficient, characteristic path length, and global efficiency, further enable the assessment of functional integration across frequency bands during resting-state\u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e. Together, these EEG-derived signatures form a multidimensional framework for characterizing both state-dependent and trait-like neural mechanisms underlying gaming addiction.\u003c/p\u003e\u003cp\u003eIn the present study, we aim to identify EEG biomarkers associated with IGD by jointly analyzing event-related EEG responses to in-game events and resting-state brain connectivity across frequency bands. we analyzed ERP responses related to in-game slay and defeat events, including both time-domain and time\u0026ndash;frequency components, as well as brain connectivity features (WPLI, clustering coefficient, path length, and global efficiency) from eyes-closed and eyes-open resting-state EEG. We examined their correlations with IGD-20 scores and subscales, and further applied canonical correlation analysis (CCA) to quantify the relationship between ERP and resting-state features and IGD severity. We identified defeated-ERPs and eyes-closed delta-band connectivity features are correlated with IGD. These findings suggest that both stable brain network organization and context-specific neural responses jointly shape individual vulnerability to gaming addiction.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cb\u003eSelf-reported gaming behavior and IGD-20 scores\u003c/b\u003e\u003c/p\u003e\u003cp\u003eTo assess gaming addiction, both subjective and objective indicators were evaluated. The subjective measure consisted of self-reported weekly gaming hours for \u003cem\u003eHonor of Kings\u003c/em\u003e, which varied widely across subjects (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eA). As an objective measure, the IGD-20 was used to assess six dimensions of gaming disorder: salience, mood modification, tolerance, withdrawal, conflict, and relapse. Total IGD-20 scores showed substantial variability across subjects (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eB), with distinct distributions observed across the six subscales (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eC to H). To examine the relationship between subjective gaming behavior and gaming disorder severity, we conducted correlation analyses between weekly gaming hours and IGD-20 total scores. The results revealed no significant association between these two measures (Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e), suggesting that self-reported gaming duration may not reliably reflect the severity of gaming addiction.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003eDefeated-induced cerebral responses are associated with gaming addiction\u003c/b\u003e\u003c/p\u003e\u003cp\u003eTo investigate real-time neural responses during gameplay, we analyzed ERPs and time-frequency dynamics in response to two types of in-game events: slay (the subject slayed an opponent) and defeated (the subject was defeated by an opponent). In the time domain, both event types elicited a positive ERP component peaking between 300 and 400 ms post-event, with topographic maps indicating maximal activity over frontal-parietal regions (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA and S1). Time-frequency analysis revealed alpha-band event-related desynchronization between 600 and 1000 ms post-event for both event types, and a clear beta-band event-related synchronization only following the defeated events between 300 and 1000 ms (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eB). Statistical comparisons showed that the ERP amplitude in the defeated condition was significantly larger than in the slay condition (p\u0026thinsp;\u0026lt;\u0026thinsp;0.0001; Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eC). Beta-band power was significantly greater following defeated events than slay events (p\u0026thinsp;\u0026lt;\u0026thinsp;0.0001; Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eD). However, no significant difference was observed in alpha-band power between the two conditions (p\u0026thinsp;=\u0026thinsp;0.4368; Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eE).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eWe next examined the correlations between IGD-20 scores and all EEG signatures (i.e., ERP amplitudes, beta power, and alpha power) for both slay and defeated events (Table S2). Pearson correlation analyses revealed that only defeated-related neural responses were significantly associated with IGD-20 scores (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). Specifically, alpha-band power was positively correlated with the IGD total score (r\u0026thinsp;=\u0026thinsp;0.4609, p\u0026thinsp;=\u0026thinsp;0.0269; Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA), conflict score (r\u0026thinsp;=\u0026thinsp;0.5028, p\u0026thinsp;=\u0026thinsp;0.0145; Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB), and salience score (r\u0026thinsp;=\u0026thinsp;0.4597, p\u0026thinsp;=\u0026thinsp;0.0273; Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eC). In contrast, ERP amplitude was negatively correlated with the relapse subscale (r = -0.4555, p\u0026thinsp;=\u0026thinsp;0.0290; Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eD), and beta-band power was positively correlated with the conflict subscale (r\u0026thinsp;=\u0026thinsp;0.4520, p\u0026thinsp;=\u0026thinsp;0.0303; Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eE). These results suggest that neural responses to defeated events, rather than slay events, may reflect gaming addiction.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003eResting-state brain connectivity features are related to gaming addiction\u003c/b\u003e\u003c/p\u003e\u003cp\u003eResting-state EEG has been widely employed to characterize neurophysiological signatures of gaming disorder\u003csup\u003e\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u003c/sup\u003e. While previous studies have primarily focused on spectra\u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e, micro state\u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u003c/sup\u003e, and coherence analysis\u003csup\u003e\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e, we extended this analysis by examining functional connectivity using the WPLI and derived graph-theoretical metrics, including clustering coefficient, average path length, and global efficiency. Figure\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e presents resting-state EEG brain connectivity features under both eyes-open and eyes-closed conditions across five frequency bands. Notably, in the alpha and beta bands, the eyes-closed condition showed significantly higher whole-brain WPLI (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA; Table S3A), clustering coefficient (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eB; Table S3B), and global efficiency (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eD; Table S3D), along with significantly lower mean path length compared to the eyes-open condition (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eC; Table S3C). These results suggest that brain networks are more integrated and efficient during the eyes-closed resting state.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eWe next examined associations between IGD-20 scores and resting-state brain connectivity features (Table S4). Notably, alpha-band connectivity in the eyes-open condition was consistently correlated with the conflict subscale of the IGD-20, suggesting that alpha-band network features may serve as reliable neural markers of gaming-related interpersonal conflict (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e). In contrast, delta-band connectivity in the eyes-closed condition showed widespread associations with IGD severity. Specifically, WPLI was positively correlated with the IGD total score, tolerance, withdrawal symptoms, and relapse (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eA); clustering coefficient was positively correlated with the total score, tolerance, and relapse (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eB); mean path length was negatively correlated with the total score, mood modification, and relapse (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eC); and global efficiency was positively correlated with the total score, withdrawal symptoms, and relapse (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eD). These findings suggest that low-frequency functional integration during rest is associated with multiple behavioral dimensions of gaming addiction. In addition, several scattered correlations were observed, including an association between gaming hours and delta-band clustering coefficient in the eyes-open condition (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eB), between IGD total score and relapse with theta-band mean path length in the eyes-closed condition (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eC), and between conflict score and beta-band global efficiency in the eyes-open condition (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eD). Collectively, these results highlight the importance of resting-state connectivity, particularly in the delta frequency bands, as potential electrophysiological markers of gaming addiction.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003eCanonical correlation between neural features and gaming disorder\u003c/b\u003e\u003c/p\u003e\u003cp\u003eTo examine the multivariate relationship between EEG features and IGD severity, we conducted a CCA (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e). Based on prior findings, we selected delta-band brain connectivity features from the eyes-closed resting-state condition, i.e., WPLI, clustering coefficient, mean path length, and global efficiency, as representative measures of intrinsic functional integration, given their consistent associations with IGD scores. In parallel, we selected defeated-related ERP features, including amplitude, beta-band power, and alpha-band power within their ROIs, as representative task-evoked neural responses. To reduce dimensionality, principal component analysis (PCA) was performed separately on the resting-state and ERP features. The first principal component (PC1) from each set was extracted, yielding two orthogonal neural features. These two neural features were then entered into the CCA model to examine their joint relationship with IGD severity, as measured by the total IGD-20 score.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eCCA revealed a significant canonical correlation between neural features and IGD score (r\u0026thinsp;=\u0026thinsp;0.662, p\u0026thinsp;=\u0026thinsp;0.0031, Wilks\u0026rsquo; Lambda), indicating a strong multivariate association. Canonical loadings indicated that delta-band brain connectivity contributed more strongly to the canonical variate (loading\u0026thinsp;=\u0026thinsp;0.876), while the defeated-ERP component contributed moderately (loading = -0.522). A bootstrap analysis with 1,000 iterations confirmed the robustness of this relationship, yielding a 95% confidence interval ranging from 0.441 to 0.863. These results suggest that IGD severity is reliably associated with a combined neural signature encompassing both resting-state connectivity and event-related EEG responses.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eThe present study integrated psychometric assessments, task-related EEG, and resting-state brain connectivity analyses to investigate the neurophysiological underpinnings of gaming addiction in adolescents and young adults. By leveraging a naturalistic MOBA gameplay paradigm and a publicly available EEG dataset, we explored both evoked neural responses to in-game events and intrinsic functional network properties during rest. Our findings revealed that electrophysiological features, particularly those elicited during defeat events and resting-state delta-band connectivity, were significantly associated with the severity of IGD-20. CCA further demonstrated a robust multivariate relationship between these neural markers and IGD scores, suggesting that gaming addiction is reflected in a combined neural signature encompassing both reactive and intrinsic brain processes.\u003c/p\u003e\u003cp\u003e\u003cb\u003eSubjective self-reported gaming hour may have bias\u003c/b\u003e\u003c/p\u003e\u003cp\u003eAn eccentric observation from our analysis is the lack of correlation between self-reported weekly gaming hours and IGD-20 total scores. The gaming hours are often used as a proxy for gaming addiction\u003csup\u003e\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u003c/sup\u003e, while our findings suggest that this measure may not accurately capture the psychological dimensions of gaming addiction. One possible explanation is that self-reported gaming time was subject to bias, particularly social desirability bias\u003csup\u003e\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e,\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u003c/sup\u003e. During participation in formal experiments, especially those involving neural recordings or addiction-related assessments, participants may intentionally underreport their gaming duration in an effort to present themselves in a more favorable light. Additionally, retrospective recall inaccuracies may further compromise the reliability of self-reported time estimates. In contrast, standardized psychometric instruments like the IGD-20 capture the broader emotional, cognitive, and behavioral features of gaming disorder, providing a more stable and comprehensive assessment\u003csup\u003e\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u003c/sup\u003e. This discrepancy underscores the importance of incorporating validated diagnostic scales rather than relying solely on subjective behavioral metrics when evaluating gaming-related pathology.\u003c/p\u003e\u003cp\u003e\u003cb\u003eDefeated-induced non-phase locking oscillations are sensitive to gaming disorder\u003c/b\u003e\u003c/p\u003e\u003cp\u003eOur task-based EEG analyses revealed that defeat-related neural responses, rather than those associated with slay, were more strongly correlated with IGD severity. In particular, alpha- and beta-band power were positively correlated with gaming disorder severity, and ERP amplitude was negatively related to relapse score. The non-phase locking oscillations, especially the beta-ROI, possible is related to cognitive and emotional processing\u003csup\u003e\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u003c/sup\u003e, suggesting the stronger emotional responses to negative event predicts a serious gaming disorder. These findings align with prior studies suggesting that individuals with gaming addiction exhibit heightened emotional reactivity to negative outcomes, worth emotional regulation, and impaired cognitive control in response to in-game stressors\u003csup\u003e\u003cspan additionalcitationids=\"CR19 CR20\" citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u003c/sup\u003e. Notably, these event-related EEG markers may provide an ecologically valid window into how individuals with IGD process failure and loss during competitive gameplay.\u003c/p\u003e\u003cp\u003e\u003cb\u003eBrain connectivity features potentially serve as the biomarkers of gaming addiction\u003c/b\u003e\u003c/p\u003e\u003cp\u003eAlpha-band connectivity features during the eyes-open resting state (including whole-brain WPLI, clustering coefficient, and global efficiency) were significantly associated with the conflict subscale of the IGD-20. This subscale measures the extent to which gaming behavior disrupts interpersonal relationships, academic performance, or occupational functioning. Conceptually, it reflects impairments in self-regulation, social-emotional monitoring, and executive control in real-world contexts\u003csup\u003e\u003cspan additionalcitationids=\"CR23\" citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e. alpha-band network activity is associated with cognitive control, information integration, conflict monitoring, and emotional regulation\u003csup\u003e\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e,\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u003c/sup\u003e. Therefore, the observed negative associations suggest that individuals with higher levels of gaming-related conflict exhibit weakened alpha-band network integration during eyes-open rest. This finding supports the interpretation that reduced alpha-band connectivity may reflect impairments in externally directed attention and social-emotional regulation, particularly in contexts involving interpersonal or functional conflict.\u003c/p\u003e\u003cp\u003eIn contrast, delta-band connectivity metrics during the eyes-closed resting state showed consistent with multiple behavioral dimensions of IGD. Delta-band components are usually associated with large-scale neural integration\u003csup\u003e\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e\u003c/sup\u003e, and are often implicated in neuropsychiatric conditions such as addiction and depression\u003csup\u003e\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e,\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e\u003c/sup\u003e. In this study, individuals with higher IGD severity exhibited greater delta-band global efficiency and clustering coefficient, along with shorter average path length, suggesting a more compact and integrated network configuration. This compensatory integration may reflect a neuroadaptive reorganization of brain systems involved in motivation, emotional regulation, and cognitive control as a result of extensive gaming exposure. Given that eyes-closed resting-state EEG captures more intrinsic, trait-like brain activity\u003csup\u003e\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u003c/sup\u003e, delta-band functional connectivity may serve as a reliable and stable neurophysiological biomarker of gaming addiction.\u003c/p\u003e\u003cp\u003eTaken together, the distinct yet complementary patterns observed in alpha and delta bands across different resting-state conditions suggest frequency-specific and state-dependent neural correlates of IGD. While alpha-band alterations may relate more to attention and social conflict regulation in externally engaged states, delta-band connectivity reflects broader intrinsic changes in large-scale network integration. Future studies should test the predictive validity of these band-specific features in larger and longitudinal cohorts and explore their responsiveness to interventions such as cognitive control training or neurofeedback therapy.\u003c/p\u003e\u003cp\u003e\u003cb\u003eSelection and dimensionality reduction of neural features for CCA\u003c/b\u003e\u003c/p\u003e\u003cp\u003eTo examine the joint neural correlates of gaming addiction, we conducted a CCA using electrophysiological features drawn from two domains: event-related EEG responses and resting-state connectivity. Specifically, we selected defeated-related ERP features and delta-band resting-state connectivity for inclusion in the model. This operation was based on prior univariate analyses showing that these two neural domains demonstrated the strongest and most consistent associations with IGD severity. Defeated-related EEG features were significantly correlated with various IGD-20 subscales, particularly relapse and conflict. Likewise, delta-band connectivity metrics from the eyes-closed resting state showed robust correlations with multiple IGD dimensions. These findings collectively indicated that both reactive neural responses to in-game failure and intrinsic low-frequency network organization are tightly linked to the behavioral manifestations of gaming addiction.\u003c/p\u003e\u003cp\u003eGiven the relatively small sample size (n\u0026thinsp;=\u0026thinsp;23) and the high dimensionality of the input features, we applied PCA to each neural domain to reduce the number of variables and improve the robustness of the CCA estimation. From each domain, only the first principal component was retained, capturing the dominant variance within defeated-related ERP features and delta-band connectivity features, respectively. This dimensionality reduction strategy minimizes the risk of overfitting and ensures stable estimation of canonical variates in small-sample multivariate models\u003csup\u003e\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e,\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u003c/sup\u003e. The resulting two neural components were then used to represent the neural dataset in the CCA.\u003c/p\u003e\u003cp\u003e\u003cb\u003eLimitation\u003c/b\u003e\u003c/p\u003e\u003cp\u003eThis study has two primary limitations that should be considered when interpreting the findings. First, although the sample size (n\u0026thinsp;=\u0026thinsp;23) yielded acceptable statistical power (approximately 0.71 under the assumptions of r\u0026thinsp;=\u0026thinsp;0.5 and α\u0026thinsp;=\u0026thinsp;0.05), it remains modest and limits the generalizability of the results. The small sample size also constrained our ability to apply more advanced analytical methods, such as structural equation modeling or machine learning-based classification approaches, which typically require larger datasets to ensure model stability and prevent overfitting\u003csup\u003e\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e,\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e\u003c/sup\u003e. Future studies with larger and more diverse samples would allow for more comprehensive modeling of the complex relationships between neural dynamics, psychological traits, and gaming behavior.\u003c/p\u003e\u003cp\u003eSecond, the cross-sectional design of this study prevents causal inference regarding the directionality of the observed brain\u0026ndash;behavior associations. The identified defeated-related ERP responses and delta-band connectivity features may either reflect predisposing neural risk markers for Internet Gaming Disorder or emerge as neuroplastic adaptations resulting from prolonged gaming. Longitudinal or intervention-based designs that monitor electrophysiological changes over time are necessary to disentangle these possibilities and to determine the predictive utility of the identified neural features.\u003c/p\u003e\u003cp\u003eThird, the dataset used in this study was drawn exclusively from Chinese young adults, a population with a distinct cultural context regarding gaming norms, social expectations, and education systems\u003csup\u003e\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e\u003c/sup\u003e. Cultural factors are known to influence both the prevalence and expression of gaming disorder\u003csup\u003e\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e\u003c/sup\u003e, as well as the neural and behavioral responses associated with gaming experiences. Therefore, caution is warranted in generalizing these findings to other populations. Future studies should aim to replicate and extend the current results across more diverse cultural and demographic groups to examine the cross-cultural robustness and universality of the identified EEG markers.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eThe present study provides novel evidence linking both task-evoked and resting-state electrophysiological features to the severity of IGD-20 in young individuals. We demonstrated that neural responses to defeat during naturalistic gameplay, specifically ERP amplitude and post-event beta and alpha power—were significantly associated with key dimensions of gaming addiction, including relapse and conflict. Additionally, resting-state connectivity in the delta frequency band, particularly under eyes-closed conditions, was strongly correlated with multiple IGD subscales, suggesting its potential as a stable neural marker of intrinsic functional organization in addiction. Multivariate analysis via canonical correlation further revealed that a joint neural signature combining defeated-related ERP features and delta-band connectivity robustly predicted IGD severity. Together, these findings highlight the utility of integrating time-locked and intrinsic EEG features to capture the multifaceted neural underpinnings of gaming addiction and offer promising targets for future diagnostic and intervention efforts.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003e\u003cb\u003eParticipants and procedure\u003c/b\u003e\u003c/p\u003e\u003cp\u003e23 healthy subjects (8 female and 15 male participants), aged from 15 to 22 years, were recruited for this study. Each subject reported their average weekly gaming hours and completed the Internet Gaming Disorder Scale-20 (IGD-20). Both the total score and six subscale scores (salience, mood modification, tolerance, withdrawal, conflict, and relapse) were recorded for further analysis. Subjects were instructed to play 5 to 7 rounds of the multiplayer online battle arena (MOBA) game \u003cem\u003eHonor of Kings\u003c/em\u003e under naturalistic conditions\u003csup\u003e\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003e\u003cb\u003eEEG recording\u003c/b\u003e\u003c/p\u003e\u003cp\u003eElectroencephalography (EEG) data were obtained from a publicly available dataset published by Li et al. on OpenNeuro\u003csup\u003e\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e\u003c/sup\u003e. EEG signals were recorded using a 64-channel Neuroscan system at a sampling rate of 1000 Hz. Two types of EEG data were acquired: (1) gameplay recordings with event markers for game start, game end, slaying an enemy, and being defeated by an enemy; and (2) resting-state recordings (eyes-open and eyes-closed) lasting approximately 6 to 7 minutes.\u003c/p\u003e\u003cp\u003e\u003cb\u003eEEG preprocessing\u003c/b\u003e\u003c/p\u003e\u003cp\u003eEEG preprocessing was conducted using EEGLAB in MATLAB\u003csup\u003e\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u003c/sup\u003e. Both game-playing and resting-states EEG signals were band-pass filtered between 1 and 45 Hz and re-referenced to the average of the M1 and M2 electrodes. Noisy channels, identified as having power spectral values more than three standard deviations from the mean, were removed and interpolated using spherical spline interpolation. Independent component analysis (ICA) was used to remove ocular and muscle artifacts, following procedures\u003csup\u003e\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u003c/sup\u003e. For the gameplay data, epochs were extracted from − 500 ms to 1500 ms relative to two event markers: \"slay\" (when the participant slayed an enemy) and \"defeated\" (when the participant was defeated by an enemy). Baseline correction was performed using the pre-event window (-500 to 0 ms). For resting-state EEG, a 200-second segment from the middle of the recording was selected for subsequent analysis.\u003c/p\u003e\u003cp\u003e\u003cb\u003eEvent-related EEG components extraction\u003c/b\u003e\u003c/p\u003e\u003cp\u003eEvent-related potentials (ERPs) were obtained by averaging the epoched data filtered from 1 to 30 Hz. ERPs were computed over a region of interest (ROI) comprising nine frontal-parietal electrodes (F1, Fz, F2, FC1, FCz, FC2, C1, Cz, and C2). For each condition, the ERP peak amplitudes were extracted as the maximum value occurring between 200 and 400 ms post-stimulus. Time-frequency distributions (TFDs) were computed using MATLAB’s built-in function \u003cem\u003espectrogram\u003c/em\u003e, with a 200-ms Hanning window, 190 ms overlap of widow, and 1000 nfft, yielding a time resolution of 10 ms and frequency resolution of 1 Hz. Baseline correction was applied to each trial by normalizing the TFDs using the − 200 to 0 ms pre-stimulus interval. Beta-band power (14 to 30 Hz) was averaged over the ROI during 300 to 1000 ms post-event, and alpha-band power (8 to 13 Hz) was averaged over 600 to 1000 ms.\u003c/p\u003e\u003cp\u003e\u003cb\u003eResting-state EEG brain connectivity features extraction\u003c/b\u003e\u003c/p\u003e\u003cp\u003eResting-state EEG data were analyzed using the Brain Connectivity Toolbox in MATLAB\u003csup\u003e\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u003c/sup\u003e. Signals were filtered into five frequency bands: delta (1 to 4 Hz), theta (4 to 8 Hz), alpha (8 to 13 Hz), beta (13 to 30 Hz), and gamma (31 to 45 Hz). For each band, the weighted phase lag index (WPLI) was computed between all pairs of 60 electrodes to estimate phase synchronization while minimizing volume conduction\u003csup\u003e\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u003c/sup\u003e. The mean of all pairwise wPLI values was used to represent whole-brain functional connectivity\u003csup\u003e\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eTo characterize the topological properties of the brain network, three graph-theoretical metrics were derived from the wPLI matrices: clustering coefficient\u003csup\u003e\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e\u003c/sup\u003e, average path length\u003csup\u003e\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e\u003c/sup\u003e, and global efficiency\u003csup\u003e\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e\u003c/sup\u003e. Distance matrices were calculated as the inverse of the WPLI matrices (with appropriate handling of zeros and infinities), and used to compute shortest paths. The clustering coefficient was averaged across nodes, while average path length and global efficiency were computed from the resulting distance matrices.\u003c/p\u003e\u003cp\u003e\u003cb\u003eCanonical correlation analysis\u003c/b\u003e\u003c/p\u003e\u003cp\u003eTo examine the relationship between gaming disorder and neural indices, a canonical correlation analysis (CCA) was conducted in R using the \u003cem\u003eCCP\u003c/em\u003e package. The gaming disorder consisted of the IGD total score, while the neural variable set included electrophysiological measures derived from principal component analysis (PCA) of defeated-related components (ERP amplitude, beta ROI power, alpha ROI power) and delta-band brain connectivity features (whole brain WPLI, clustering coefficient, average path length, and global efficiency). Specifically, PCA was first applied separately to defeated-related components and delta-band brain connectivity features. The first principal component (PC1) of defeated ERP features (ERP_PC1) and the first PC of delta band brain connectivity features (delta_PC1) were extracted and used as summary indices. These two components represented the neural data matrix \u003cem\u003eY\u003c/em\u003e, while the IGD total score served as the behavioral matrix \u003cem\u003eX\u003c/em\u003e. Both \u003cem\u003eX\u003c/em\u003e and \u003cem\u003eY\u003c/em\u003e were standardized prior to analysis. CCA was then performed to identify the maximal linear relationship between brain activities and gaming disorder. Canonical loadings, defined as the Pearson correlation coefficients between the original neural variables (ERP_PC1 and delta_PC1) and the corresponding canonical variate, were also calculated to assess variable contributions.\u003c/p\u003e\u003cp\u003e\u003cb\u003eStatistical analyses\u003c/b\u003e\u003c/p\u003e\u003cp\u003eAll statistical analyses were conducted using GraphPad Prism (version 9) and jamovi (version 2.3). Paired-sample \u003cem\u003et\u003c/em\u003e tests were used to compare ERP amplitude, alpha ROI power, and beta ROI power between the slay and defeated conditions. Brain connectivity features were analyzed using two-way analysis of variance (ANOVA) with Bonferroni post hoc correction. Pearson correlation and linear regression analyses were performed to examine relationships between IGD-20 scores, gameplay behavior, and EEG-derived features. For the CCA results, statistical significance of the canonical correlation coefficient was evaluated using Wilks’ lambda approximation. To assess the robustness of the first canonical correlation, a non-parametric bootstrap analysis with 1000 iterations was conducted by resampling with replacement. The resulting bootstrap distribution was used to compute the 95% confidence interval (CI), defined by the 2.5th and 97.5th percentiles.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cspan\u003eEthics Statement This study was approved by the Ethics Committee of University-Town Hospital of Chongqing Medical University (Approval No. LL-202307). The EEG data analyzed in this manuscript were obtained from a previously published, open-access dataset (Li et al., 2024, OpenNeuro accession number: ds005520; Li et al., 2025, Scientific Data). All procedures involving human participants in the original study were conducted in accordance with the ethical standards of the institutional research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. Participant Consent In the original study from which the EEG data used in this manuscript were obtained (Li et al., 2024, OpenNeuro accession number: ds005520; Li et al., 2025, Scientific Data), all participants were fully informed about the study procedures, potential risks, and their rights prior to enrollment. Written informed consent to participate in the study and for the subsequent use and publication of their anonymized data was obtained from each participant under the oversight of the Ethics Committee of University-Town Hospital of Chongqing Medical University (Approval No. LL-202307). For the current secondary analysis, no new human data were collected, and the authors only accessed and analyzed de-identified, open-access data.\u003c/span\u003e\u003c/p\u003e\u003cp\u003e\u003cstrong\u003eAcknowledgements\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe thank Hong-Zhi Li, Jia-Jia Yang, Zhen Lv, Li-Yang Wan, Wo Wang, Da-Qi Li, Dong-Dong Zhou, ang Li Kuang for the data support.\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding declaration:\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors received no specific funding for this work.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor contributions:\u003c/strong\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eFeng-Rui Zhang designed study. Jonathan M. Dong and Feng-Rui Zhang analyzed the data and wrote the manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConflicts of interest\u003c/strong\u003e:\u003c/p\u003e\n\u003cp\u003eNone.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData and script availability statement\u003c/strong\u003e:\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe dataset used in this study is publicly available on OpenNeuro (https://openneuro.org/datasets/ds005520/versions/1.0.0). All analysis scripts are available from the corresponding author upon reasonable request.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eMora-Cantallops, M. \u0026amp; Sicilia, M.-\u0026Aacute;. MOBA games: A literature review. \u003cem\u003eEntertainment Computing\u003c/em\u003e \u003cstrong\u003e26\u003c/strong\u003e, 128-138, doi: 10.1016/j.entcom.2018.02.005 (2018).\u003c/li\u003e\n\u003cli\u003eCui, L., Kim, E. J. \u0026amp; Kim, J. How Chinese Media Addresses Esports Issues: A Text Mining Comparative Analysis of Online News and Viewers\u0026rsquo; Comments on the Hangzhou Asian Games. \u003cem\u003eElectronics\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, 4961 (2023).\u003c/li\u003e\n\u003cli\u003ePontes, H. M., Rumpf, H.-J., Selak, \u0026Scaron;. \u0026amp; Montag, C. Investigating the interplay between gaming disorder and functional impairments in professional esports gaming. \u003cem\u003eSci Rep-Uk\u003c/em\u003e \u003cstrong\u003e14\u003c/strong\u003e, 6557, doi:10.1038/s41598-024-56358-x (2024).\u003c/li\u003e\n\u003cli\u003eWischert-Zielke, M. \u0026amp; Barke, A. Differences between recreational gamers and Internet Gaming Disorder candidates in a sample of Animal Crossing: New Horizons players. \u003cem\u003eSci Rep-Uk\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, 5102, doi:10.1038/s41598-023-32113-6 (2023).\u003c/li\u003e\n\u003cli\u003eGao, Y.-X., Wang, J.-Y. \u0026amp; Dong, G.-H. The prevalence and possible risk factors of internet gaming disorder among adolescents and young adults: Systematic reviews and meta-analyses. \u003cem\u003eJournal of Psychiatric Research\u003c/em\u003e \u003cstrong\u003e154\u003c/strong\u003e, 35-43, doi: 10.1016/j.jpsychires.2022.06.049 (2022).\u003c/li\u003e\n\u003cli\u003ePetry, N. M.\u003cem\u003e et al.\u003c/em\u003e An international consensus for assessing internet gaming disorder using the new DSM-5 approach. \u003cem\u003eAddiction\u003c/em\u003e \u003cstrong\u003e109\u003c/strong\u003e, 1399-1406, doi: 10.1111/add.12457 (2014).\u003c/li\u003e\n\u003cli\u003eZhang, J.-T.\u003cem\u003e et al.\u003c/em\u003e Altered resting-state functional connectivity of the insula in young adults with Internet gaming disorder. \u003cem\u003eAddiction Biology\u003c/em\u003e \u003cstrong\u003e21\u003c/strong\u003e, 743-751, doi: 10.1111/adb.12247 (2016).\u003c/li\u003e\n\u003cli\u003ePark, S., Ha, J., Ahn, W. \u0026amp; Kim, L. Measurement of craving among gamers with internet gaming disorder using repeated presentations of game videos: a resting-state electroencephalography study. \u003cem\u003eBMC Public Health\u003c/em\u003e \u003cstrong\u003e23\u003c/strong\u003e, 816, doi:10.1186/s12889-023-15750-4 (2023).\u003c/li\u003e\n\u003cli\u003eWang, Z.\u003cem\u003e et al.\u003c/em\u003e Cortical complexity and connectivity during isoflurane-induced general anesthesia: a rat study. \u003cem\u003eJournal of Neural Engineering\u003c/em\u003e \u003cstrong\u003e19\u003c/strong\u003e, 036009, doi:10.1088/1741-2552/ac6a7b (2022).\u003c/li\u003e\n\u003cli\u003eBurleigh, T. L., Griffiths, M. D., Sumich, A., Wang, G. Y. \u0026amp; Kuss, D. J. Gaming disorder and internet addiction: A systematic review of resting-state EEG studies. \u003cem\u003eAddictive Behaviors\u003c/em\u003e \u003cstrong\u003e107\u003c/strong\u003e, 106429, doi: 10.1016/j.addbeh.2020.106429 (2020).\u003c/li\u003e\n\u003cli\u003eKim, J. S.\u003cem\u003e et al.\u003c/em\u003e Resting-state EEG microstate analysis of internet gaming disorder and alcohol use disorder. \u003cem\u003eDialogues in Clinical Neuroscience\u003c/em\u003e \u003cstrong\u003e26\u003c/strong\u003e, 89-102, doi:10.1080/19585969.2024.2432913 (2024).\u003c/li\u003e\n\u003cli\u003eLee, J.-Y., Choi, J.-S. \u0026amp; Kwon, J. S. Neurophysiological Mechanisms of Resilience as a Protective Factor in Patients with Internet Gaming Disorder: A Resting-State EEG Coherence Study. \u003cem\u003eJournal of Clinical Medicine\u003c/em\u003e \u003cstrong\u003e8\u003c/strong\u003e, 49 (2019).\u003c/li\u003e\n\u003cli\u003eFuster, H., Carbonell, X., Pontes, H. M. \u0026amp; Griffiths, M. D. Spanish validation of the Internet Gaming Disorder-20 (IGD-20) Test. \u003cem\u003eComputers in Human Behavior\u003c/em\u003e \u003cstrong\u003e56\u003c/strong\u003e, 215-224, doi: 10.1016/j.chb.2015.11.050 (2016).\u003c/li\u003e\n\u003cli\u003eKrumpal, I. Determinants of social desirability bias in sensitive surveys: a literature review. \u003cem\u003eQuality \u0026amp; Quantity\u003c/em\u003e \u003cstrong\u003e47\u003c/strong\u003e, 2025-2047, doi:10.1007/s11135-011-9640-9 (2013).\u003c/li\u003e\n\u003cli\u003eKing, M. F. \u0026amp; Bruner, G. C. Social desirability bias: A neglected aspect of validity testing. \u003cem\u003ePsychology \u0026amp; Marketing\u003c/em\u003e \u003cstrong\u003e17\u003c/strong\u003e, 79-103, doi: 10.1002/(SICI)1520-6793(200002)17:2\u0026lt;79::AID-MAR2\u0026gt;3.0.CO;2-0 (2000).\u003c/li\u003e\n\u003cli\u003eShu M., Y., Ivan Jacob, A. P., Meng Xuan, Z. \u0026amp; Anise M. S., W. Psychometric validation of the Internet Gaming Disorder-20 Test among Chinese middle school and university students. \u003cem\u003eJournal of Behavioral Addictions\u003c/em\u003e \u003cstrong\u003e8\u003c/strong\u003e, 295-305, doi: 10.1556/2006.8.2019.18 (2019).\u003c/li\u003e\n\u003cli\u003eG\u0026uuml;ntekin, B. \u0026amp; Başar, E. Event-related beta oscillations are affected by emotional eliciting stimuli. \u003cem\u003eNeuroscience Letters\u003c/em\u003e \u003cstrong\u003e483\u003c/strong\u003e, 173-178, doi: 10.1016/j.neulet.2010.08.002 (2010).\u003c/li\u003e\n\u003cli\u003eYen, J.-Y.\u003cem\u003e et al.\u003c/em\u003e Emotional Regulation in Young Adults with Internet Gaming Disorder. \u003cem\u003eInternational Journal of Environmental Research and Public Health\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, 30 (2018).\u003c/li\u003e\n\u003cli\u003eEstupi\u0026ntilde;\u0026aacute;, F. J., Bernaldo-de-Quir\u0026oacute;s, M., Vallejo-Ach\u0026oacute;n, M., Fern\u0026aacute;ndez-Arias, I. \u0026amp; Labrador, F. Emotional regulation in Gaming Disorder: A systematic review. \u003cem\u003eThe American Journal on Addictions\u003c/em\u003e \u003cstrong\u003e33\u003c/strong\u003e, 605-620, doi: 10.1111/ajad.13621 (2024).\u003c/li\u003e\n\u003cli\u003eDong, G., Li, H., Wang, L. \u0026amp; Potenza, M. N. Cognitive control and reward/loss processing in Internet gaming disorder: Results from a comparison with recreational Internet game-users. \u003cem\u003eEuropean Psychiatry\u003c/em\u003e \u003cstrong\u003e44\u003c/strong\u003e, 30-38, doi:10.1016/j.eurpsy.2017.03.004 (2017).\u003c/li\u003e\n\u003cli\u003eWang, Z.\u003cem\u003e et al.\u003c/em\u003e Diversiform Etiologies for Post-stroke Depression. \u003cem\u003eFrontiers in Psychiatry\u003c/em\u003e \u003cstrong\u003eVolume 9 - 2018\u003c/strong\u003e, doi:10.3389/fpsyt.2018.00761 (2019).\u003c/li\u003e\n\u003cli\u003eGriffiths, M. D. A \u0026lsquo;components\u0026rsquo; model of addiction within a biopsychosocial framework. \u003cem\u003eJournal of Substance Use\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, 191-197, doi: 10.1080/14659890500114359 (2005).\u003c/li\u003e\n\u003cli\u003eFaraci, P., Bottaro, R., Nasonte, G., Valenti, G. D. \u0026amp; Craparo, G. A New Short Version of Internet Gaming Disorder-20: An Exploratory Structural Equation Modeling. \u003cem\u003eClin Neuropsychiatry\u003c/em\u003e \u003cstrong\u003e20\u003c/strong\u003e, 205-218, doi:10.36131/cnfioritieditore20230305 (2023).\u003c/li\u003e\n\u003cli\u003eSprong, M. E., Griffiths, M. D., Lloyd, D. P., Paul, E. \u0026amp; Buono, F. D. Comparison of the Video Game Functional Assessment-Revised (VGFA-R) and Internet Gaming Disorder Test (IGD-20). \u003cem\u003eFrontiers in Psychology\u003c/em\u003e \u003cstrong\u003eVolume 10 - 2019\u003c/strong\u003e, doi:10.3389/fpsyg.2019.00310 (2019).\u003c/li\u003e\n\u003cli\u003eWu, S.\u003cem\u003e et al.\u003c/em\u003e Disrupted Functional Brain Network Architecture in Sufferers with Boxing-Related Repeated Mild Traumatic Brain Injury: A Resting-State EEG Study. \u003cem\u003eJIN\u003c/em\u003e \u003cstrong\u003e23\u003c/strong\u003e, doi:10.31083/j.jin2305102 (2024).\u003c/li\u003e\n\u003cli\u003eLejko, N., Larabi, D. I., Herrmann, C. S., Aleman, A. \u0026amp; Ćurčić-Blake, B. Alpha Power and Functional Connectivity in Cognitive Decline: A Systematic Review and Meta-Analysis. \u003cem\u003eJournal of Alzheimer\u0026rsquo;s Disease\u003c/em\u003e \u003cstrong\u003e78\u003c/strong\u003e, 1047-1088, doi:10.3233/jad-200962 (2020).\u003c/li\u003e\n\u003cli\u003eSchwartzmann, B.\u003cem\u003e et al.\u003c/em\u003e Resting-state EEG delta and alpha power predict response to cognitive behavioral therapy in depression: a Canadian biomarker integration network for depression study. \u003cem\u003eSci Rep-Uk\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, 8418, doi:10.1038/s41598-023-35179-4 (2023).\u003c/li\u003e\n\u003cli\u003eNewson, J. J. \u0026amp; Thiagarajan, T. C. EEG Frequency Bands in Psychiatric Disorders: A Review of Resting State Studies. \u003cem\u003eFrontiers in Human Neuroscience\u003c/em\u003e \u003cstrong\u003eVolume 12 - 2018\u003c/strong\u003e, doi:10.3389/fnhum.2018.00521 (2019).\u003c/li\u003e\n\u003cli\u003eCostumero, V., Bueichek\u0026uacute;, E., Adri\u0026aacute;n-Ventura, J. \u0026amp; \u0026Aacute;vila, C. Opening or closing eyes at rest modulates the functional connectivity of V1 with default and salience networks. \u003cem\u003eSci Rep-Uk\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, 9137, doi:10.1038/s41598-020-66100-y (2020).\u003c/li\u003e\n\u003cli\u003eSinganamalli, A.\u003cem\u003e et al.\u003c/em\u003e Cascaded Multi-view Canonical Correlation (CaMCCo) for Early Diagnosis of Alzheimer\u0026rsquo;s Disease via Fusion of Clinical, Imaging and Omic Features. \u003cem\u003eSci Rep-Uk\u003c/em\u003e \u003cstrong\u003e7\u003c/strong\u003e, 8137, doi:10.1038/s41598-017-03925-0 (2017).\u003c/li\u003e\n\u003cli\u003eHelmer, M.\u003cem\u003e et al.\u003c/em\u003e On the stability of canonical correlation analysis and partial least squares with application to brain-behavior associations. \u003cem\u003eCommunications Biology\u003c/em\u003e \u003cstrong\u003e7\u003c/strong\u003e, 217, doi:10.1038/s42003-024-05869-4 (2024).\u003c/li\u003e\n\u003cli\u003eWolf, E. J., Harrington, K. M., Clark, S. L. \u0026amp; Miller, M. W. Sample Size Requirements for Structural Equation Models:An Evaluation of Power, Bias, and Solution Propriety. \u003cem\u003eEducational and Psychological Measurement\u003c/em\u003e \u003cstrong\u003e73\u003c/strong\u003e, 913-934, doi:10.1177/0013164413495237 (2013).\u003c/li\u003e\n\u003cli\u003eLi, J., Horiguchi, Y. \u0026amp; Sawaragi, T. in \u003cem\u003e2020 59th Annual Conference of the Society of Instrument and Control Engineers of Japan (SICE).\u003c/em\u003e 826-831.\u003c/li\u003e\n\u003cli\u003eLiao, Z.\u003cem\u003e et al.\u003c/em\u003e Prevalence of Internet Gaming Disorder and Its Association With Personality Traits and Gaming Characteristics Among Chinese Adolescent Gamers. \u003cem\u003eFrontiers in Psychiatry\u003c/em\u003e \u003cstrong\u003eVolume 11 - 2020\u003c/strong\u003e, doi:10.3389/fpsyt.2020.598585 (2020).\u003c/li\u003e\n\u003cli\u003eChew, Peter K. H., Lin, Patrick K. F. \u0026amp; Yow, Yong J. Cross-Cultural Differences in the Pathways to Internet Gaming Disorder. \u003cem\u003eAsia-Pacific Psychiatry\u003c/em\u003e \u003cstrong\u003e16\u003c/strong\u003e, e12565, doi: 10.1111/appy.12565 (2024).\u003c/li\u003e\n\u003cli\u003eLi, H.-Z.\u003cem\u003e et al.\u003c/em\u003e (OpenNeuro, 2024).\u003c/li\u003e\n\u003cli\u003eDelorme, A. \u0026amp; Makeig, S. EEGLAB: an open source toolbox for analysis of single-trial EEG dynamics including independent component analysis. \u003cem\u003eJournal of neuroscience methods\u003c/em\u003e \u003cstrong\u003e134\u003c/strong\u003e, 9-21, doi:10.1016/j.jneumeth.2003.10.009 (2004).\u003c/li\u003e\n\u003cli\u003eZhang, F.\u003cem\u003e et al.\u003c/em\u003e Cross-Species Investigation on Resting State Electroencephalogram. \u003cem\u003eBrain Topography\u003c/em\u003e \u003cstrong\u003e32\u003c/strong\u003e, 808-824, doi:10.1007/s10548-019-00723-x (2019).\u003c/li\u003e\n\u003cli\u003eRubinov, M. \u0026amp; Sporns, O. Complex network measures of brain connectivity: Uses and interpretations. \u003cem\u003eNeuroImage\u003c/em\u003e \u003cstrong\u003e52\u003c/strong\u003e, 1059-1069, doi: 10.1016/j.neuroimage.2009.10.003 (2010).\u003c/li\u003e\n\u003cli\u003eImperatori, L. S.\u003cem\u003e et al.\u003c/em\u003e Cross-participant prediction of vigilance stages through the combined use of wPLI and wSMI EEG functional connectivity metrics. \u003cem\u003eSleep\u003c/em\u003e \u003cstrong\u003e44\u003c/strong\u003e, doi:10.1093/sleep/zsaa247 (2020).\u003c/li\u003e\n\u003cli\u003eOmurtag, A.\u003cem\u003e et al.\u003c/em\u003e Disruption of functional network development in children with prenatal Zika virus exposure revealed by resting-state EEG. \u003cem\u003eSci Rep-Uk\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, 6346, doi:10.1038/s41598-025-90860-0 (2025).\u003c/li\u003e\n\u003cli\u003eMehraram, R., Kries, J., De Clercq, P., Vandermosten, M. \u0026amp; Francart, T. EEG reveals brain network alterations in chronic aphasia during natural speech listening. \u003cem\u003eSci Rep-Uk\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, 2441, doi:10.1038/s41598-025-86192-8 (2025).\u003c/li\u003e\n\u003cli\u003eGuo, H.\u003cem\u003e et al.\u003c/em\u003e A hybrid critical channels and optimal feature subset selection framework for EEG fatigue recognition. \u003cem\u003eSci Rep-Uk\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, 2139, doi:10.1038/s41598-025-86234-1 (2025).\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"gaming addiction, MOBA game, EEG, resting-state, ERP","lastPublishedDoi":"10.21203/rs.3.rs-7044193/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7044193/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eInternet Gaming Disorder (IGD) is a growing behavioral addiction among adolescents and young adults, yet its neurophysiological underpinnings remain unclear. This study aimed to identify EEG biomarkers of IGD by integrating event-related potentials (ERPs) during naturalistic gameplay with resting-state brain connectivity. We analyzed EEG responses to in-game defeat and slay events in a multiplayer online battle arena (MOBA) game and computed connectivity metrics (i.e., weighted phase lag index, clustering coefficient, path length, and global efficiency) across frequency bands during eyes-closed and eyes-open resting-state conditions. Correlation analyses revealed that neural responses to defeat, particularly beta-band synchronization and ERP amplitudes, were significantly associated with IGD severity. Additionally, delta-band connectivity during eyes-closed rest showed robust associations with multiple IGD subscales. Canonical correlation analysis further demonstrated a significant multivariate relationship between a combined neural signature and IGD severity. These findings suggest that both reactive neural responses to negative in-game events and intrinsic functional connectivity jointly contribute to individual vulnerability to gaming addiction. This multidimensional EEG framework offers novel insights into the neurobiological mechanisms of IGD and holds promise for improving assessment and intervention strategies.\u003c/p\u003e","manuscriptTitle":"Cerebral responses to in-game failure and resting-state connectivity are jointly associated with gaming addiction","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-09-19 06:54:52","doi":"10.21203/rs.3.rs-7044193/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"editorInvitedReview","content":"","date":"2026-04-14T10:14:59+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"40721087345369688309674198302192911187","date":"2026-04-14T09:04:31+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-11-06T08:51:40+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-11-04T13:09:12+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2025-07-15T06:57:00+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-07-08T08:33:53+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2025-07-08T08:30:42+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"d50481ae-c6cc-4906-83a2-0c33ed076fc3","owner":[],"postedDate":"September 19th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":53614908,"name":"Biological sciences/Neuroscience"},{"id":53614909,"name":"Biological sciences/Psychology"},{"id":53614910,"name":"Social science/Psychology"}],"tags":[],"updatedAt":"2025-11-06T08:53:35+00:00","versionOfRecord":[],"versionCreatedAt":"2025-09-19 06:54:52","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7044193","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7044193","identity":"rs-7044193","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}