Social media applications like Twitter, Facebook, and Instagram have remained consistently popular worldwide. Previous studies have focused on the risks of problematic use of social media and its associated psychological issues, including anxiety, depression, and boredom proneness (Kuss and Griffiths, 2017, Lopes et al., 2022, Camerini et al., 2023). However, most research compares overall scores of these variables and overlooks specific symptoms (e.g., tolerance or salience in behavioural addiction). Therefore, it is important to further investigate these topics using analytic methods such as network analysis. Furthermore, cross-cultural studies have identified significant cultural or national differences in social media and internet usage behaviours, particularly between Eastern and Western countries within the individualism-collectivism framework (LaRose et al., 2014, Cheng et al., 2021). However, online behaviours in collectivist countries (e.g., China and Japan) also warrant further investigation, given their different social systems and internet environments (Wu et al., 2022). Therefore, the present study aims to compare problematic social media use (PSMU) and its related psychological distress and boredom proneness in China and Japan, using network analysis, which can identify the connections and importance of symptoms from a network-based or systemic perspective. It is important to emphasise that PSMU is a particular form of PIU that centres on social networking platforms.

Internet addiction (IA) or problematic internet use (PIU) has been extensively studied since the last decade of the 20th century (Griffiths, 1996, Young, 1998). Since Davis (2001) introduced the classification of generalised and specific pathological internet use, research has concentrated on the problematic or addictive use of specific online functions of applications, such as social media (Kuss and Griffiths, 2017, Lopes et al., 2022). However, only (internet) gaming disorder and gambling disorder have been included in official diagnostic manuals such as the fifth edition (text revision) of the Diagnostic and Statistical Manual of Mental Disorders (DSM-5 TR; American Psychiatric Association, 2022) and the 11th revision of the International Classification of Diseases ICD-11 (World Health Organization, 2019). It is also important to avoid overpathologising everyday behaviours as addictions, since daily activities might not reach the severity of addiction (Billieux et al., 2015). Therefore, this study uses the term PSMU, which can be defined as the uncontrolled and excessive use of social media leading to functional impairment (Brand et al., 2019, Lopes et al., 2022).

Beyond the discussions of the concepts of PIU and PSMU, studies focus on the associations or impacts, such as depression and anxiety. The association between PSMU and psychological distress has been well documented in theoretical and empirical studies (Brand et al., 2019, Chen et al., 2020, Lopes et al., 2022). The Interaction of Person-Affect-Cognition-Execution (I-PACE) model suggests that psychopathological variables are the risk factors of addictive behaviours (Brand et al., 2019). According to the I-PACE model, addictive behaviours develop through interactions between person-related factors (e.g., personality traits and psychopathology), affective and cognitive responses, and executive functions (e.g., inhibitory control) (Brand et al., 2019). Within this framework, psychological distress (e.g., depression) may heighten reliance on social media as a coping mechanism, thereby reinforcing problematic or addictive use. It suggests that psychological distress is a key vulnerability factor for PSMU and provides a useful lens for interpreting the associations examined in the present study. Besides, the Compensatory Internet Use Theory (CIUT) suggests that individuals would be craving for more internet usage for compensation and to alleviate their negative feelings in negative life situations (Kardefelt-Winther, 2014). Many empirical studies suggest that PSMU is associated with higher psychological distress, such as anxiety and depression (Hussain et al., 2020, Lopes et al., 2022), which could become a longitudinal vicious loop (Twigg et al., 2020, Marttila et al., 2021, Casale et al., 2025).

Besides the theoretical and empirical evidence on the link between PSMU and psychological distress, the relationship between PIU and boredom proneness has been proven in empirical studies (Camerini et al., 2023, Tagliaferri et al., 2025). A meta-analysis reported a medium-to-large positive relationship (r = 0.342) between boredom and problematic digital media use (Camerini et al., 2023). However, in Camerini et al.’s (2023) review, five aspects of problematic digital media use were examined, including gaming, the internet, smartphones, social media, and technology in general. PSMU clearly sets itself apart from other specific PIUs that demand investigations with a particular focus on social media use. It is believed that individuals use social media to alleviate boredom, and PSMU can become boring, which can result in a vicious loop (Stockdale & Coyne, 2020). Studies have identified the association between PSMU and boredom proneness (Whelan et al., 2020, Bai et al., 2021, Malik et al., 2024, Yoosefi et al., 2025). For example, it was revealed that trait boredom predicted higher problematic Facebook use, which in turn predicted higher state boredom (Donati et al., 2022).

Therefore, based on the reviews above, individuals may use the internet or social media problematically to alleviate their psychological distress and compensate for boredom. It is necessary to investigate the link between PSMU, distress, and boredom further. However, most of those studies calculated the relationships between the total scores of PSMU, psychological distress, and boredom proneness and ignored the specific symptoms of each construct. Their findings have limited implications for more specific and symptom-targeted prevention strategies for PSMU.

The network analysis approach investigates how symptoms connect by viewing psychological phenomena as a network of interconnected nodes (Borsboom, 2017). It effectively identifies core symptoms—those highly connected nodes that are crucial to the network (Borsboom & Cramer, 2013). Network comparison tests can identify differences in symptom association patterns across different groups or developmental stages (van Borkulo et al., 2015) and evaluate the relative importance of symptoms, offering a more comprehensive understanding of symptom networks in diverse populations (Benfer et al., 2018).

Some studies have employed network analysis to examine the interconnectedness of PSMU symptoms (Svicher et al., 2021, Wang et al., 2022), and some other network studies investigated the PSMU-distress comorbidity (Wang et al., 2022, Peng and Liao, 2023, Tullett-Prado et al., 2023, Ding et al., 2024, Bai et al., 2025, Liu et al., 2026). For instance, “salience” was identified as the core symptom in the PSMU network among Chinese participants (Wang et al., 2022). Similarly, in a study among Italian participants, the most central symptom of PSMU was difficulty controlling social media use and obsessive thoughts about going online (Svicher et al., 2021). However, Li et al. (2023) reported that “tolerance” was the most central PSMU symptom. Although network analysis was widely utilised to investigate PSMU, different measures of PSMU were employed, such as the Bergen Social Media Addiction Scale (e.g., Wang et al., 2022, Peng and Liao, 2023) or the adopted 15-item scale for generalised problematic internet use for the use of social media sites (e.g., Svicher et al., 2021). It would be necessary to examine the PSMU symptom network further using the same scale in a different context and compare the results with previous findings.

Regarding comorbidity networks, varied results have been reported across studies. Among participants of different ethnicities, “tolerance” and “mood modification” were the most central symptoms in the PSMU-psychological distress network when total scores of psychological distress variables were included (Tullett-Prado et al., 2023). Peng & Liao (2023) found that “agitated” (stress), “panic” (anxiety), and “sad mood” (depression) were the key symptoms in the PSMU-psychological distress network within a Chinese sample, using individual symptoms rather than total scores. Another Chinese study identified that “(lack of) Enthusiasm” (depression) was central in the network linking PSMU, anxiety, and depression (Wang et al., 2022). A longitudinal network analysis indicates that “withdrawal-depression” was the strongest link in the PSMU-mental health networks over time, and earlier depression was the most influential predictor of subsequent PSMU symptoms in the cross-lagged panel network model (Liu et al., 2026). However, Liu et al. (2026) used the total score of depression in the network, and it remains unknown which specific depressive symptom served as the predictor of PSMU. Therefore, it is crucial to identify the most influential symptoms within the PSMU network and explore the relationship between PSMU and psychological distress symptoms through network analysis.

Furthermore, boredom (proneness) is seldom incorporated into network analysis studies of PSMU, although studies have emphasised the close link between boredom and IA (as discussed previously) (Camerini et al., 2023). One study demonstrated that loneliness was the most central node in the network, encompassing gaming disorder, depression, alexithymia, boredom, and loneliness (Li et al., 2021). However, they only used the total scores of the boredom and loneliness scales in the network and did not examine the specific symptoms. It should be noted that gaming disorder differs from PSMU, and such results may not serve as direct evidence of a link between PSMU and boredom. It is therefore important to explore the role of boredom, particularly the specific symptoms, in the network of PSMU, psychological distress, and boredom proneness.

PSMU shows notable cultural differences (Tang et al., 2018, Cheng et al., 2021, Ong and Lee, 2022). A meta-analysis of “social media addiction” across 32 nations and a cross-cultural study in seven countries suggest that collectivist cultures, including China and Japan, exhibit higher levels of PSMU, compared to individualist cultures (Tang et al., 2018, Cheng et al., 2021). Asian students had higher levels of PSMU than their peers in the USA (Tang et al., 2018). Another cross-national study of PSMU among 30 countries also indicates that participants from a collectivist society had higher PSMU levels compared with those in the individualist countries (Thomas et al., 2022). Such differences may stem from the emphasis on social connectedness in East Asian collectivistic cultures, where social media serves as a primary platform for maintaining interpersonal relationships (Tang et al., 2018). Studies tended to explain such differences using the individualism-collectivism framework (LaRose et al., 2014).

Most cross-cultural or cross-national comparisons within collectivist countries tended to focus on PIU instead of PSMU. The cross-national study mentioned above did not compare PSMU within collectivist countries (Thomas et al., 2022). Differences in problematic technology use were identified within collectivist countries, for instance, between China and Japan (Yang et al., 2013, Wu et al., 2022). For example, Japanese students tend to report higher internet gaming disorder tendencies, PIU, and more severe depression, while Chinese students exhibit more prominent anxiety symptoms (Yang et al., 2013, Wu et al., 2022), indicating culture-specific mechanisms that warrant further investigation. However, PIU or IA are different from PSMU. Differences in PIU cannot be direct evidence of differences in PSMU. To our knowledge, no study has compared PSMU between China and Japan, although PIU and gaming disorder differ between the two countries. It would be necessary to compare PSMU within collectivist countries, such as China and Japan.

China and Japan, both in the East Asian Confucian cultural circle (Tucker, 2018), share collectivist roots but differ in individualism (Hofstede et al., 2010). Existing studies focus more on comparing Asia and the West (Yang et al., 2013, Thomas et al., 2022), neglecting their nuanced differences. Their comparison can uncover subtle cultural mechanisms behind internet-related addictions and offer a more refined basis for enhancing cross-cultural understanding of PIU (Wu et al., 2022). Beyond the general collectivist framework, China and Japan have distinct digital ecosystems, which may result in different PSMU use patterns. China has several unique applications, such as WeChat and Weibo, which are not as globally popular as they are in China (Hou et al., 2018). The internet environment in Japan differs from that in China, while the widely used social media platforms include both local sites and those popular in Western countries, such as Twitter/X (Wang, 2016). Therefore, it is essential to compare PSMU between the two countries, taking into account their different digital social media environments.

While most studies have focused on differences in prevalence and correlates, to our knowledge, no study has used network analysis to compare the connections among PSMU symptoms and the comorbidity networks of PSMU and its correlates between China and Japan. It is therefore necessary to adopt network analysis to uncover cultural specificity in symptom interactions (e.g., core symptoms of PSMU), which can be crucial for effective prevention of PSMU in both countries.

Recent studies have examined the relationship between PSMU and psychological distress using network analysis. However, most of these studies did not include specific symptoms in their networks but instead used total scores, especially for the psychological distress variables (e.g., Li et al., 2023, Tullett-Prado et al., 2023, Liu et al., 2026). Studies using total scores of the scales assume that all items equally reflect a single underlying construct. However, studies in network psychometrics have shown that aggregating all items can obscure the heterogeneity of behavioural and emotional symptoms (Borsboom, 2017). A symptom-level network approach treats each symptom as an active, interacting element rather than a passive indicator of a latent variable. This method enables the identification of specific central symptoms that are crucial in sustaining behavioural patterns (Bringmann et al., 2019, Epskamp et al., 2018). Therefore, symptom-level network analysis can identify more precise mechanisms of comorbidity and reveal intervention targets that are hidden when using aggregated scale scores. Besides, the role of boredom proneness symptoms is rarely explored when investigating the networks of PSMU and psychological distress. Therefore, it is essential to conduct network analysis to understand how PSMU symptoms are connected and how they relate to psychological distress and boredom proneness symptoms. Another research gap is that cross-cultural or cross-national network analysis remains limited in examining PSMU and psychological distress. Studies have explored and identified cultural differences in PSMU (Tang et al., 2018, Cheng et al., 2021). However, cross-cultural comparisons of the networks of PSMU and psychological distress symptoms are scarce. It is crucial to examine these associations across China and Japan, given their shared collectivist roots but differing social systems, communication practices, and platform environments (such as WeChat/TikTok in China and LINE/Twitter in Japan). Accordingly, the present study has several research aims: (1) to estimate the network structures of PSMU symptoms in China and Japan; (2) to estimate the network structures of PSMU, boredom proneness, and psychological distress (including anxiety, depression, and stress) symptoms in both countries; (3) to identify the central or influential nodes within the networks in China and Japan; and (4) to compare the networks between the two countries.