Abstract
Internet gaming disorder is an increasing public health problem due to the widespread availability of online gaming. Social media platforms drive this trend by enabling gameplay sharing and increasing user engagement, potentially reinforcing addictive gaming behaviors. Understanding how gaming content exposure on social media affects brain activity in individuals with internet gaming disorder is crucial. This study aimed to investigate gaming content neural responses on social media in individuals with internet gaming disorder using functional magnetic resonance imaging. We aimed to determine differences in activation patterns that contribute to understanding the neurobiological underpinnings of internet gaming disorder by examining brain activity in these individuals and comparing it to healthy controls. Additionally, we investigated the association of brain activity with clinical characteristics (internet gaming disorder severity and illness duration). The participants with internet gaming disorder demonstrated increased bilateral orbitofrontal cortex, bilateral hippocampus, left precuneus, and right superior temporal gyrus activation in response to gaming-related cues on social media compared to healthy controls. Additionally, internet gaming disorder severity and illness duration correlated with left hippocampus activation levels. These results improve our understanding of how gaming-related content on social media affects individuals with internet gaming disorder. Our findings provide valuable information into the neurobiological features of internet gaming disorder and help develop effective treatment interventions.
Introduction
Internet gaming disorder (IGD) has recently appeared as a significant public health issue, coinciding with the immense popularity of online games (King et al. 2020; Zhang et al. 2020). IGD is characterized by the excessive and compulsive use of internet games, causing negative consequences in various aspects of an individual’s life, including mental health, social relationships, and academic performance (Niu et al. 2022; Xue et al. 2025). Hence, several studies have been conducted. However, effective treatments remain limited, and cases frequently become challenging to manage. One contributing factor is the pervasive presence of the internet and smartphones in daily life, which often exposes individuals with IGD to audiovisual stimuli that trigger their addiction, causing repeated exacerbations of addictive behavior (Kuss and Griffiths 2017; Ko et al. 2020; Jitoku et al. 2024; Kobayashi et al. 2024). Addressing this issue has become urgent with the coronavirus disease 2019 pandemic accelerating the shift toward online activities.
The accessibility of online gaming has expanded in recent years, particularly through social media platforms (eg TikTok, Instagram, YouTube), which have become major sources of gaming-related content. These platforms offer a gameplay medium and provide venues for sharing and consuming game-related information, reviews, and live streams (Carr and Hayes 2015; Balakrishnan and Griffiths 2017; Kuss and Griffiths 2017). They significantly affect user engagement and potentially reinforce gaming-related addictive behaviors (Kuss and Griffiths 2017; Fujimoto et al. 2024; Jitoku et al. 2024). Investigating the association of exposure to gaming content on social media with the brain activity of individuals with IGD is crucial, considering the immersive and interactive nature of social media and gaming.
Functional magnetic resonance imaging (fMRI) is a powerful tool for assessing the neural correlates of behavioral and psychological phenomena (Dong et al. 2017; Wang et al. 2017; Takeuchi et al. 2022). Our previous fMRI study investigated the neural activity induced by gaming-related content on social media among young adults who casually played online games (Fujimoto et al. 2024). While undergoing fMRI assessment, the participants viewed gaming-related videos and neutral (non-gaming) videos on social media. We found that gaming-related cues caused significant activation of several key brain regions, including the medial prefrontal cortex (MPFC), posterior cingulate cortex (PCC), hippocampus, thalamus, superior temporal gyrus (STG), and precuneus, compared to neutral cues. These findings indicate that these brain regions play a crucial role in how gaming content on social media influences neural activity. However, as our previous study (Fujimoto et al. 2024) did not include participants with IGD, the specific impact of such content on individuals with the disorder remains unclear.
Cue reactivity plays a crucial role in the development and maintenance of addictive behaviors, and numerous studies have examined the brain mechanisms underlying cue reactivity using fMRI. A substantial body of evidence indicates that brain regions involved in reward, memory, and attention processing play important roles in substance and behavioral addictions (Jasinska et al. 2014; Starcke et al. 2018). Along this line, several studies have explored brain responses to gaming cues in individuals with IGD. For example, previous studies have shown that when individuals with IGD viewed gaming-related images, an increase in signal activity in brain areas such as the MPFC, anterior cingulate cortex (ACC), orbitofrontal cortex (OFC), and hippocampus was observed (Sun et al. 2012; Wang et al. 2017; Yu et al. 2021). Dong et al. (2017) found that in individuals with IGD, gaming led to heightened craving and increased activation in the MPFC, striatum, and precuneus when comparing data collected before and after gaming. However, the majority of previous studies have used images or videos of familiar games as stimuli without contextual information. Hence, neural responses to gaming-related content on social media among individuals with IGD remain unknown. Social media offers a real-world setting where gaming-related content is encountered alongside diverse social and environmental influences, such as interactive posts, advertisements, and user-generated content (Carr and Hayes 2015; Kuss and Griffiths 2017). This dynamic environment indicates how individuals engage with gaming content in their daily lives, enabling us to assess neural responses in an ecologically valid setting. Building on this perspective, comparing individuals with IGD to healthy casual gamers can help identify core elements specific to IGD, distinguishing it from non-problematic gaming. While gaming is an enjoyable activity for many, the majority of players can regulate their gaming behavior without significant difficulties (Dong et al. 2017; Niu et al. 2022). However, only a small subset develops IGD, highlighting the need to understand the factors that contribute to its emergence. Based on the findings of previous fMRI studies on cue reactivity in IGD (Sun et al. 2012; Dong et al. 2017; Wang et al. 2017; Yu et al. 2021), we hypothesized that, when exposed to gaming content on social media, IGD participants would exhibit greater activity than healthy casual gamers in the brain regions associated with reward processing, memory, and attentional control, including the prefrontal areas, hippocampus, precuneus, and STG.
Moreover, neuroimaging parameters have been linked to clinical measures across various types of addiction. For instance, a previous study observed that hippocampal and amygdalar volumes were correlated with behavioral inhibition system in individuals with gambling disorder (Rahman et al. 2014). More recently, Huntley et al. (2020) reported that increased resting-state connectivity between the ventral striatum and the hippocampus predicts higher levels of substance use in adolescents. Remarkably, hippocampal parameters have repeatedly been reported to be associated with clinical characteristics in individuals with IGD. For example, the volumes in the hippocampus were positively correlated with severity of IGD symptoms (Yoon et al. 2017). Additionally, Jeong et al. (2016) identified increased fractional anisotropy in a broad white matter region including the right cingulum to the hippocampus, which correlated with illness duration in IGD subjects. Frequent exposure to gaming-related stimuli may reinforce neural pathways central to the hippocampus, leading to more ingrained habits and a heightened likelihood of compulsive behavior (Castilla-Ortega et al. 2016; Qiu et al. 2023). These findings suggest that the hippocampus, a region in the brain involved in memory retrieval and experiential learning, plays a vital role in the clinical characteristics of IGD and that hippocampal parameters could serve as valuable prognostic markers for IGD. Given that gaming content on social media platforms—such as videos of individuals enjoying a game, introducing a game, or teaching game strategies—contains a real-world context where gaming-related content is experienced along with various social and environmental factors, exposure to this content could trigger the retrieval of related memories and learned strategies from past gaming experiences. Therefore, we hypothesized that hippocampal activity levels induced by gaming content on social media would correlate with clinical characteristics, such as IGD severity and illness duration, in individuals with IGD.
This study aimed to examine the neural responses to gaming content on social media in individuals with IGD using fMRI. We determined potential differences in activation patterns that could help understand the neurobiological underpinnings of IGD by examining the brain activity in these individuals and comparing it to healthy controls who casually played online games. Furthermore, we explored the association of brain activity with clinical characteristics, including IGD severity and illness duration, among the participants with IGD.
Materials and methods
Participants
This study recruited 17 male patients diagnosed with IGD (aged 18 to 38 yr, mean ± SD = 21.5 ± 5.1 yr), according to the criteria proposed by the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5). Specifically, these patients met at least five of the nine inclusion criteria outlined in the DSM-5. Additionally, current psychiatric comorbidity was assessed using the structured clinical interview for DSM-5. Moreover, 2, 3, 2, and 1 patients were diagnosed with autism spectrum disorder (ASD), attention-deficit/hyperactivity disorder (ADHD), ASD and ADHD, and ADHD and obsessive-compulsive disorder (OCD), respectively. Among them, 1, 1, 2, 1, and 1 received ADHD medication, sleep medication, both ADHD and sleep medications, antipsychotic medication, and both selective serotonin reuptake inhibitor and sleep medication, respectively. None of the participants reported any history of head trauma, severe medical or surgical illness, or substance abuse. According to previous studies (Dong et al. 2018, 2021), the severity of IGD was estimated using the number of DSM-5 criteria met by participants (mean ± SD = 6.7 ± 1.4). The illness duration was 3.6 ± 2.9 yr. The mean intelligence quotient (IQ) estimated with the Japanese version of the National Adult Reading Test short form (Matsuoka et al. 2006) was 107.6 ± 8.3. The data from age- and IQ-matched healthy male participants analyzed in the previous study (Fujimoto et al. 2024) were utilized to configure the healthy control group (HC group, n = 22, age 21.5 ± 2.2 yr, IQ 107.3 ± 7.2). Supplementary Methods describes the details.
The institutional review board of Institute of Science Tokyo Hospital approved this study which conformed to the Code of Ethics of the World Medical Association (ethical approval number R-2021-006). All participants signed written informed consent after receiving an explanation of the entire study.
fMRI task
The task was identical to that used in our previous study (Fujimoto et al. 2024). Please refer to Supplementary Methods for details. In brief, we selected eight gaming-related videos from social media platforms, featuring individuals enjoying a game, introducing a game, or providing tutorials on game capture. Each video showcased a different game. Additionally, we selected eight neutral (non-gaming) videos from social media, covering topics such as furniture, hygiene, travel, and work. Following previous studies (Tapert et al. 2003; Vollstädt-Klein et al. 2011; Chen et al. 2018), these neutral videos were chosen to closely match the gaming-related ones in terms of content, complexity, design, color, luminance, and the presence of faces.
Each video lasted 20 s and was presented in a pseudorandomized order. Participants rated their gaming desire on a scale from 1 (no desire) to 4 (extreme desire) within 6 s after watching each video (Fig. 1). To assess gaming desire levels, we calculated the difference between the mean gaming desire ratings for gaming-related and neutral videos, with higher scores indicating greater gaming desire. Unfortunately, the scores of two participants with IGD were not recorded due to technical errors. Outside the scanner, the participants were asked to rate each game’s familiarity presented in the fMRI task from 1 (very unfamiliar) to 9 (very familiar).
fMRI task. Videos (20 s each) were pseudorandomly displayed, and participants were requested to rate their gaming desire from 1 (no desire) to 4 (extreme desire) after watching each video. An illustrative frame from a gaming-related video used in the task is shown for reference.
fMRI data acquisition and preprocessing
All participants underwent MRI scanning using a 3-T whole-body scanner (Prisma; Siemens, Erlangen, Germany) equipped with a 20-channel head/neck coil. SPM12 (Wellcome Trust Center for Neuroimaging, London, UK) in MATLAB (MathWorks, Natick, MA, USA) was used to process images. Supplementary Methods presents further details.
Data analysis
Demographic and behavioral data
SPSS version 21 was used for demographic and behavioral data analyses. Some of our continuous measures did not follow a normal distribution (Shapiro–Wilk test, P < 0.05). Therefore, we used the Mann–Whitney test to compare group differences and Spearman’s rank correlation for correlation analyses.
fMRI data
We applied a general linear model to the fMRI data. The design matrix for the first-level analysis included two task-related regressors of interest: gaming-related/neutral content conditions. Additionally, we incorporated six movement parameters as regressors of no interest (three displacements and three rotations).
In the second-level analyses, we employed a factorial design to examine the main effects of group and condition, as well as the interaction between group and condition. Based on prior literature (Dong et al. 2017; Liu et al. 2017; Ma et al. 2019; Yu et al. 2021) and in line with our previous study (Fujimoto et al. 2024), we focused on specific regions of interest (ROIs), including the MPFC, orbitofrontal cortex (OFC), middle frontal gyrus (MFG), ACC, PCC, striatum, thalamus, hippocampus, STG, and precuneus. Anatomical masks for these ROIs were obtained from the Automated Anatomical Labeling atlas (Tzourio-Mazoyer et al. 2002; Maldjian et al. 2003). Significance was determined using family-wise error (FWE) correction for multiple comparisons, with a cluster-level threshold of P < 0.01 for each ROI (voxel-level uncorrected P < 0.001). Additionally, for exploratory analysis, we reported activations that met a voxel-level threshold of P < 0.01 (FWE corrected) with a minimum cluster extent of 100 contiguous voxels, following whole-brain correction for multiple comparisons in line with previous studies (Sami et al. 2023; Tei et al. 2023).
Spearman’s rank correlations were calculated between the clinical characteristics (IGD severity [DSM-5 score] and illness duration) and the parameter estimates (game-related condition–neutral condition), which were extracted as the first eigenvariate from the clusters demonstrating significant group-by-condition interactions in the IGD group (no correction was applied for multiple comparisons).
Results
Participants with IGD demonstrated a stronger gaming desire in response to gaming-related cues on social media compared to neutral cues (2.48 ± 0.70 vs. 1.27 ± 0.49 [Wilcoxon test: Z = −3.35, P < 0.01, effect size r = 0.61]). Although the levels of gaming desire were numerically higher in the IGD group compared with the HC group, no significant differences were observed between the groups (HC 1.05 ± 0.59, IGD 1.21 ± 0.78 [Mann–Whitney test: U = 142.00, Z = −0.71, P = 0.49, effect size r = 0.12]). Similar to the HC group (Fujimoto et al. 2024), the levels of gaming desire showed no significant correlation with familiarity rating scores in the IGD group (P = 0.68). Regarding the response time, there were no significant differences between the gaming-related and neutral conditions in both groups (see Supplementary Results for details).
The main effect of the condition revealed that the bilateral MPFC, bilateral MFG, bilateral OFC, bilateral ACC, bilateral PCC, left striatum, bilateral thalamus, bilateral hippocampus, right STG, and bilateral precuneus demonstrated significantly greater activation in response to gaming-related cues on social media compared to neutral ones for the ROI analysis. Conversely, no significant regions exhibited deactivation (neutral cues > gaming-related cues) at the same statistical thresholds. The main effect of group showed no significant activation. Details are described in Table S1. However, the group-by-condition interaction indicated significant differences between groups in the bilateral OFC, bilateral hippocampus, right STG, and left precuneus (Fig. 2 and Table 1). As illustrated in Fig. 2, these differences were mainly caused by increased activation in response to gaming-related cues in participants with IGD. In these regions, the activation levels (game-related condition–neutral condition) in the left hippocampus were positively correlated with IGD severity (rho = 0.51, df = 15, P = 0.037) and illness duration (rho = 0.49, df = 15, P = 0.048) (Fig. 3). The activation levels in other brain regions demonstrated no significant correlations with either IGD severity or illness duration among participants with IGD (all, P > 0.05).
Brain regions group-by-condition interaction. The error bars indicate ± standard errors; P < 0.01 and cluster-level FWE corrected (at voxel-level uncorrected P < 0.001). Abbreviations: L = left, OFC = orbitofrontal cortex, R = right, STG = superior temporal gyrus.
fMRI results (interaction effect of the group-by-condition [ROI analyses]).
| Brain region | Coordinates (mm) | F | Cluster (voxels) | ||
|---|---|---|---|---|---|
| x | y | z | |||
| L orbitofrontal cortex | −4 | 46 | −22 | 35.32 | 131 |
| L hippocampus | −34 | −22 | −16 | 26.51 | 266 |
| L precuneus | −6 | −48 | 54 | 23.45 | 234 |
| R orbitofrontal cortex | 32 | 30 | −8 | 18.17 | 150 |
| R hippocampus | 34 | −20 | −16 | 19.42 | 158 |
| R superior temporal gyrus | 52 | −48 | 22 | 22.03 | 118 |
| Brain region | Coordinates (mm) | F | Cluster (voxels) | ||
|---|---|---|---|---|---|
| x | y | z | |||
| L orbitofrontal cortex | −4 | 46 | −22 | 35.32 | 131 |
| L hippocampus | −34 | −22 | −16 | 26.51 | 266 |
| L precuneus | −6 | −48 | 54 | 23.45 | 234 |
| R orbitofrontal cortex | 32 | 30 | −8 | 18.17 | 150 |
| R hippocampus | 34 | −20 | −16 | 19.42 | 158 |
| R superior temporal gyrus | 52 | −48 | 22 | 22.03 | 118 |
P < 0.01, cluster-level FWE corrected (at voxel-level uncorrected P < 0.001). MNI coordinates and F-values were provided for the local voxel maximum of each respective cluster. Abbreviations: FWE = family-wise error, L = left, MNI = Montreal Neurological Institute, R = right, ROI = region of interest.
fMRI results (interaction effect of the group-by-condition [ROI analyses]).
| Brain region | Coordinates (mm) | F | Cluster (voxels) | ||
|---|---|---|---|---|---|
| x | y | z | |||
| L orbitofrontal cortex | −4 | 46 | −22 | 35.32 | 131 |
| L hippocampus | −34 | −22 | −16 | 26.51 | 266 |
| L precuneus | −6 | −48 | 54 | 23.45 | 234 |
| R orbitofrontal cortex | 32 | 30 | −8 | 18.17 | 150 |
| R hippocampus | 34 | −20 | −16 | 19.42 | 158 |
| R superior temporal gyrus | 52 | −48 | 22 | 22.03 | 118 |
| Brain region | Coordinates (mm) | F | Cluster (voxels) | ||
|---|---|---|---|---|---|
| x | y | z | |||
| L orbitofrontal cortex | −4 | 46 | −22 | 35.32 | 131 |
| L hippocampus | −34 | −22 | −16 | 26.51 | 266 |
| L precuneus | −6 | −48 | 54 | 23.45 | 234 |
| R orbitofrontal cortex | 32 | 30 | −8 | 18.17 | 150 |
| R hippocampus | 34 | −20 | −16 | 19.42 | 158 |
| R superior temporal gyrus | 52 | −48 | 22 | 22.03 | 118 |
P < 0.01, cluster-level FWE corrected (at voxel-level uncorrected P < 0.001). MNI coordinates and F-values were provided for the local voxel maximum of each respective cluster. Abbreviations: FWE = family-wise error, L = left, MNI = Montreal Neurological Institute, R = right, ROI = region of interest.
Clinical characteristics and activation levels in the left hippocampus in participants with IGD. Left hippocampus activation levels were positively correlated with IGD severity (rho = 0.51, P = 0.037) and illness duration (rho = 0.49, P = 0.048). Abbreviations: IGD = internet gaming disorder.
The exploratory whole-brain analysis for the main effect of condition revealed that gaming-related cues on social media elicited significantly greater activation in the bilateral cerebellum, along with widespread activation across frontal, temporal, and parietal regions, compared to neutral cues. Additionally, the right cuneus and left calcarine regions exhibited deactivation (neutral cues > gaming-related cues). No significant main effect of group or group-by-condition interaction was observed in the exploratory whole-brain analysis. Table S2 presents the details.
Discussion
This study used fMRI to investigate neural responses to gaming-related content on social media in individuals with IGD. No significant differences in the levels of self-reported gaming desire were found between the groups, but individuals with IGD demonstrated increased activation in the bilateral OFC, bilateral hippocampus, right STG, and left precuneus compared to the HC group. Additionally, IGD severity and illness duration were associated with activation levels in the left hippocampus. These results provide valuable information into the neurobiological characteristics of IGD and help develop effective treatment interventions.
As expected, we observed a significant main effect of condition, with the gaming-related condition eliciting greater activation than the neutral condition in our predefined ROIs, including the MPFC, MFG, OFC, ACC, PCC, striatum, hippocampus, thalamus, STG, and precuneus. These findings underscore the critical roles of brain regions associated with reward processing, memory, and attentional control in habit formation and addictive behaviors. Moreover, our exploratory whole-brain analysis revealed greater activation in the bilateral cerebellum. Although the cerebellum was not included in our predefined ROIs, previous studies have demonstrated its involvement in spatial processing, reinforcement learning, and reward prediction, all of which are crucial components of the gaming experience (Sun et al. 2012; Wang et al. 2022a).
Our factorial analyses (ROI-based and whole-brain) revealed no significant main group effect, indicating that, overall, the brain activity did not differ significantly between the groups. This indicates that, at a general level, both groups engaged similar neural circuits in response to the task stimuli.
Intriguingly, we revealed significant group-by-condition interaction in the bilateral OFC, bilateral hippocampus, right STG, and left precuneus. Participants with IGD demonstrated increased activation in these regions in response to gaming-related content on social media compared to HC. Additionally, activation levels in the hippocampus were correlated with the clinical characteristics (IGD severity and illness duration) among participants with IGD. These brain areas have been repeatedly altered in individuals with IGD (Niu et al. 2022; Schettler et al. 2022).
Extensive evidence suggests that the hippocampus is essential for encoding and retrieving addiction-related cues, learning from experience, and regulating motivational drives, including those associated with rewarding behaviors like gaming (Castilla-Ortega et al. 2016; Goodman and Packard 2016; Kutlu and Gould 2016; Qiu et al. 2023). Accordingly, individuals with IGD have repeatedly demonstrated altered hippocampal activity when exposed to gaming cues (Niu et al. 2022; Schettler et al. 2022). In line with these findings, the current study, using ecologically valid cues, emphasizes the importance of the hippocampus in the neurobiological underpinnings of IGD. Our previous study showed that the hippocampus was activated in casual gamers when viewing gaming-related content on social media (Fujimoto et al. 2024), suggesting that such stimuli trigger memory retrieval processes, likely linked to past gaming experiences. In this context, the current findings add to our previous studies and contribute to broader discussions on the clinical implications of hippocampal involvement in the compulsive urge to engage in habitual gaming. Consistent with research on other behavioral addictions driven by memory-related cravings (Goodman and Packard 2016; Niu et al. 2022; Schettler et al. 2022; Qiu et al. 2023), our results suggest that heightened hippocampal activation in individuals with IGD, compared to casual gamers, reflects stronger associations between gaming content and personal experiences. These reinforced connections may further solidify gaming-related habits, making them more resistant to change.
Furthermore, IGD severity and illness duration were associated with left hippocampus activation levels. Previous studies have revealed that hippocampal parameters were correlated with clinical characteristics of IGD, and the authors indicated that prolonged gaming exposure may change brain regions associated with memory and executive control, potentially contributing to excessive gaming behavior (Yoon et al. 2017; Qiu et al. 2023). Although speculative, patients with more severe IGD in the current study may demonstrate a more pronounced memory-related response to gaming cues. Similarly, the association between hippocampal activation and illness duration further indicates that the longer an individual has IGD, the stronger the memory and learning-related activation becomes, possibly reinforcing habitual gaming behaviors over time. Our results emphasize the importance of early intervention in IGD to prevent the progressive strengthening of maladaptive neural pathways. Future studies need to directly investigate these speculations using a longitudinal design.
The OFC is involved in evaluating rewards, regulating emotions, and making decisions based on potential outcome evaluations (Cox et al. 2005; Dixon et al. 2017). It plays a vital role in the pathology of both substance and behavioral addictions, including IGD, by interacting with the mesolimbic dopaminergic system (Balodis et al. 2016; Niu et al. 2022; Schettler et al. 2022). Wang et al. (2017) found that individuals with IGD showed increased OFC activation during a cue reactivity task compared to recreational users, with this activation correlating with the level of desire for gaming. Additionally, a previous PET study demonstrated a link between D2 receptor levels and glucose metabolism in the OFC of individuals with IGD, suggesting that D2 receptor–mediated OFC dysregulation contributes to compulsive behavior processes and impaired control in IGD (Tian et al. 2014). We found no significant increase in OFC activity among casual gamers in our previous study (Fujimoto et al. 2024). This indicates that, although they may find gaming-related content engaging, casual gamers do not experience the same degree of reward anticipation or craving-like responses as individuals with IGD. Altogether, the current findings suggest that gaming-related content on social media triggers heightened reward responses in individuals with IGD, potentially contributing to their compulsive gaming behavior and difficulty in resisting gaming stimuli.
The right STG showed significantly greater activation in response to gaming-related cues on social media in participants with IGD compared to HCs. The STG plays a role in audiovisual processing and social cognition (Karnath 2001; Deen et al. 2015; Yi et al. 2019; Tei et al. 2020; Tolkacheva et al. 2024), making it particularly relevant when engaging with social media content. Our previous study demonstrated that the STG was activated in casual gamers while viewing gaming-related content on social media (Fujimoto et al. 2024). The fact that both casual gamers and individuals with IGD exhibited increased STG activation while viewing gaming-related content suggests that such stimuli are socially and emotionally salient to both groups. However, the STG activation in IGD individuals was significantly greater, implying heightened responsiveness to gaming-related social cues and possibly reflecting greater emotional attachment to the gaming community or game content. Such heightened STG activation may reinforce excessive gaming behaviors, as social interactions within game environments would become particularly rewarding as a result, contributing to the compulsive engagement in games characteristic of IGD. A recent study found abnormal functional connectivity and executive dysfunction in the posterior–superior temporal sulcus of adolescents with IGD, suggesting that excessive exposure to game-related social stimuli during adolescence may disrupt the dynamic interaction between the salience and social brain networks (Lee et al. 2020). This exposure has been linked to executive dysfunction and cognitive impairments (Lee et al. 2020). Taken together, the present findings highlight the need for further research on the long-term impact of gaming-related content on social media in individuals with IGD.
Moreover, the left precuneus was significantly activated in participants with IGD. This area has been involved in cue reactivity in numerous addictions (Engelmann et al. 2012; Zeng et al. 2021). The precuneus, located in the parietal lobe, is involved in various complex functions, including self-referential processing, visuospatial imagery, and attention allocation (Zhang and Volkow 2019; Dadario and Sughrue 2023; Bertoni et al. 2024). Gaming frequently involves rich visual environments requiring players to navigate and interact spatially (Jitoku et al. 2024). In our previous research, we found that, like the hippocampus and STG, the precuneus was activated in casual gamers when they viewed gaming-related content on social media (Fujimoto et al. 2024). The activation of this region in both casual gamers and IGD individuals suggests that gaming-related content on social media effectively captures attention and elicits mental simulations of gaming experiences. However, the significantly higher precuneus activation in IGD individuals implies stronger self-referential processing and visuospatial imagery in response to gaming-related stimuli, which warrants further investigation.
The results of this study have several clinical implications. The proliferation of gaming content on social media platforms has been involved in exacerbating and potentially triggering relapses in IGD in recent years (Kuss and Griffiths 2017; Ko et al. 2020; Jitoku et al. 2024; Kobayashi et al. 2024). The current results emphasize the need for clinicians to consider the effect of social media on patients and incorporate strategies to manage exposure to such content as part of a comprehensive treatment plan. The correlation between clinical characteristics (IGD severity and illness duration) and hippocampal activation indicates that early intervention in IGD is crucial to prevent the strengthening of maladaptive neural pathways over time. Therapeutic interventions that focus on modifying memory associations and reducing the salience of gaming cues are effective in treating IGD. Clinicians can develop more effective interventions to help individuals with the disorder achieve better outcomes and improve their quality of life by focusing on the specific brain regions and cognitive functions involved in IGD and considering the effect of social media. Furthermore, neuroimaging using ecologically valid cues is considered a valuable tool in evaluating treatment efficacy and monitoring changes in brain activation patterns over the course of therapy.
This study had several limitations. First, although we closely matched each gaming-related stimulus with its neutral counterpart, using naturalistic stimuli with real-world content from social media made it challenging to achieve perfect matching across multiple parameters. Factors such as the foreground–background ratio and zooming speed remained difficult to control completely, despite our efforts to carefully select neutral stimuli. Second, 2, 3, 2, and 1 patients were diagnosed with ASD, ADHD, ASD and ADHD, and ADHD and OCD, respectively. Previous neuroimaging studies revealed that individuals with these psychiatric disorders demonstrated altered brain function in various brain areas (Tei et al. 2022; Chen et al. 2024; Cordova et al. 2024; Tamon et al. 2024). Therefore, the influence of these comorbid disorders on the current findings cannot be entirely excluded. However, these comorbidities are prevalent in IGD, indicating that these disorders may share common neurobiological features (Yen et al. 2017; Ko et al. 2020). This evidence supports the hypothesis that our IGD sample represents a typical IGD population. Third, six patients with IGD received psychotropic medications. These medications generally affect brain function in various ways (Linke et al. 2017; Del Fabro et al. 2021). Fourth, the sample size was similar to that of previous fMRI studies on IGD (Ko et al. 2013; Lee et al. 2020), but it was still relatively small. Fifth, our sample exclusively consisted of males. Previous fMRI studies have indicated possible sex differences in the mechanisms underlying gaming addiction (Dong et al. 2018; Wang et al. 2022b). Therefore, caution should be considered when generalizing our results to female patients with IGD. Sixth, in our correlation analyses, we made no corrections for the multiple comparisons between brain activity and clinical characteristics of the participants with IGD. Finally, the study’s cross-sectional design precludes any causal conclusions. A longitudinal study with a larger sample size, including both sexes, and using more sophisticated analytical approaches is warranted to confirm and strengthen the current results.
Despite these limitations, the present findings contribute to a deeper understanding of the neural effects of gaming-related content on social media in individuals with IGD. Further research using real-world gaming cues should help refine treatment interventions for this population.
Acknowledgments
The authors wish to extend their gratitude to the research team of Institute for Quantum Life Science, National Institutes for Quantum Science and Technology for their assistance in data acquisition.
Author contributions
Yuka Fujimoto (Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Writing—original draft, Writing—review & editing), Junya Fujino (Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Writing—original draft, Writing—review & editing), Daisuke Matsuyoshi (Data curation, Investigation, Supervision, Writing—review & editing), Daisuke Jitoku (Conceptualization, Data curation, Investigation, Project administration, Writing—review & editing), Nanase Kobayashi (Conceptualization, Data curation, Investigation, Writing—review & editing), Chenyu Qian (Data curation, Investigation, Validation, Writing—review & editing), Shoko Okuzumi (Data curation, Investigation, Methodology, Writing—review & editing), Shisei Tei (Methodology, Supervision, Validation, Writing—review & editing), Takehiro Tamura (Conceptualization, Investigation, Supervision, Writing—review & editing), Takefumi Ueno (Conceptualization, Funding acquisition, Investigation, Methodology, Project administration, Supervision, Writing—review & editing), Makiko Yamada (Conceptualization, Funding acquisition, Investigation, Methodology, Project administration, Supervision, Writing—review & editing), and Hidehiko Takahashi (Conceptualization, Funding acquisition, Investigation, Methodology, Project administration, Supervision, Writing—review & editing).
Funding
This work was supported by the Japan Agency for Medical Research and Development (JP23dk0307102, JP24dk0307128), JST Moonshot R&D Grant (JPMJMS2295-01), and Intramural Research Grant (4-1) for Neurological and Psychiatric Disorders of NCNP and KDDI Corporation (KDDI Research, Inc). This study was also supported in part by KAKENHI JP (23H04979) from the Ministry of Education, Culture, Sports, Science and Technology of Japan and CREST (JPMJCR22P3) from the Japan Science and Technology Agency.
Conflict of interest statement
None declared.
