Unveiling symptom-level pathways: a network analysis of fear of missing out and mobile phone addiction among Chinese college students Authors: Yang He, Ting Pan, Songyue Hong, Tao Xu, Lili Liu, Shuang Liu, Ping Xiong, Xuna Yang, Dong Wang, Wenqin Chen, Wei Liu Published: 2026 ABSTRACT BackgroundPrior studies have revealed the relationship between fear of missing out (FoMO) and mobile phone addiction based on total scores. However, there has been a lack of research exploring how the specific dimensions of mobile phone addiction and items of FoMO are interrelated. The present study aimed to investigate the structural relationships between these two components and identify critical nodes using network analysis.MethodsA sample of 1,616 Chinese college students (aged 18–25) completed the Fear of Missing Out Scale (FoMOs) and the Mobile Phone Addiction Index (MPAI). We employed network analysis to estimate the network structure and identify central and bridge symptoms using Expected Influence (EI) and Bridge Expected Influence (BEI).ResultsThe network revealed stronger connections within the FoMO and mobile phone addiction communities than between them. Beyond this, several theoretically meaningful cross-community connections were identified. Among nodes, "Feeling anxiety and lost" (FAL), "Fear of friends' superior experiences" (FoMO2), "Fear of social oblivion" (FoMO3), and "Wasting time on tracking information" (FoMO6) exhibited the highest EI values. Crucially, "Feeling anxiety and lost" (FAL) and "Wasting time on tracking information" (FoMO6) also demonstrated the highest BEI values in their respective communities, identifying them as the most influential bridge nodes.ConclusionThe connections between specific FoMO items and mobile phone addiction dimensions are crucial for understanding their fine-grained relationships. Among the identified central nodes, "Feeling anxiety and lost" (FAL) and "Wasting time on tracking information" (FoMO6) were uniquely influential, serving as both central nodes within their respective communities and the key bridges connecting them. This dual role positions "Feeling anxiety and lost" and "Wasting time on tracking information" as statistically prominent nodes that warrant further investigation. Future research is needed to test whether modifying these nodes may attenuate the co-occurrence of FoMO and mobile phone addiction.Supplementary InformationThe online version contains supplementary material available at https://doi.org/10.1186/s40359-026-05257-z. Introduction Mobile phones have become integral to college students' daily lives. On average, students spend approximately 278.67 min on their mobile phones and send about 76.68 text messages each day [1]. Such frequent use inevitably increases the risk of mobile phone addiction, a compulsive condition marked by loss of control, withdrawal symptoms, and impaired social functioning [2]. This behavior causes physical problems such as visual fatigue and tactile abnormalities, and is associated with psychological difficulties including sleep disturbances, loneliness, anxiety, and depression; in severe cases, it may contribute to extreme outcomes such as self-harm and suicide [3–5]. Owing to its high prevalence, rapid growth, and serious consequences, mobile phone addiction has become a major global public health concern [6]. This issue is particularly prominent among Chinese college students. Recent studies estimate its prevalence in this population at 36.6% [4], exceeding the global average of 26.96% [6], with a consistent upward trend over the past decade [7]. These findings underscore the urgent need to investigate the underlying mechanisms of mobile phone addiction among Chinese college students to inform effective prevention and intervention strategies. One key psychological factor implicated in mobile phone addiction is Fear of Missing Out (FoMO). FoMO is defined as a pervasive apprehension that others might be having rewarding experiences from which one is absent, characterized by a persistent desire to stay continually connected with what others are doing [8]. This apprehension drives compulsive phone checking and prolonged online engagement, which are core behaviors associated with excessive, potentially addictive mobile phone use [9]. Erikson's psychosocial development theory helps explain why the link between FoMO and mobile phone addiction may be particularly pronounced among college students [10]. During young adulthood (ages 18–30), individuals face the psychosocial task of resolving the crisis between intimacy and isolation [10], which can heighten sensitivity to social experiences. Within collectivist cultural contexts that emphasize group harmony and social integration [11], forming meaningful relationships is not only a psychological need but also an important social competence. This cultural emphasis can intensify concerns about missing social interactions. Although mobile phones provide convenient tools for maintaining relationships and may temporarily alleviate FoMO, they also increase the risk of compulsive use and addiction. Self-Determination Theory (SDT) further explains the motivational basis of FoMO [12, 13]. SDT posits that autonomy, competence, and relatedness constitute three fundamental psychological needs, and that frustration of these needs, particularly relatedness, can heighten FoMO [8]. Mobile phones may partially satisfy the need for relatedness by enabling individuals to stay informed about others' activities and maintain social connections [14]. Consistent with Compensatory Internet Use Theory (CIUT) [15], excessive online engagement may also serve as a coping mechanism for negative emotions or unmet social needs in offline life. When college students experience loneliness or FoMO, they may turn to online interactions to compensate for deficiencies in real-world social fulfillment [16]. Although this behavior may offer temporary relief, overreliance on online engagement can reinforce maladaptive patterns of use and ultimately lead to mobile phone addiction. Together, these theoretical perspectives explain how FoMO may promote excessive mobile phone use among college students. Empirical evidence strongly supports the association between FoMO and mobile phone addiction. Numerous studies have demonstrated a significant relationship between these constructs [8, 17, 18], and a meta-analysis found a moderate-to-strong correlation between FoMO and mobile phone addiction among college students [19]. The interactive features of smartphones may intensify this link. Although smartphones facilitate social connection and enhance perceived social support [14, 16], the constant stream of online updates also compels individuals to monitor others' activities to avoid social exclusion [14, 20]. This compulsive monitoring behaviour reinforces dependency on mobile devices [20], leading college students into a vicious cycle where mobile phone addiction and FoMO mutually reinforce one another [21]. However, most existing studies have examined this relationship using total-score approaches, treating FoMO and mobile phone addiction as unitary constructs [19–21]. This common practice may obscure specific relationships among individual symptoms and fail to reveal the fine-grained interplay between components of these constructs. To address this limitation, researchers have increasingly adopted network analysis, an approach that conceptualizes psychological constructs as dynamic systems of interacting symptoms [22]. As a data-driven, emerging analytical method, network analysis does not depend on a priori assumptions regarding relationships among variables [22, 23] and allows complex association patterns to be visualized intuitively through network structures [24]. Compared with traditional total-score approaches, network analysis offers three major advantages. First, it enables detailed examination of fine-grained associations among variables and facilitates exploration of potential pathways underlying these relationships [25]. Second, by calculating centrality indices, it quantifies the relative importance of individual symptoms within the network and identifies highly influential nodes [26]. Third, bridge centrality analysis enables the identification of bridge nodes that connect distinct symptom communities, which are often considered promising targets for psychological intervention [27]. This approach has demonstrated substantial value in clarifying complex relationships in addiction research. For instance, network analytic studies have identified strong interconnectedness among gambling, internet use, and social media use [28], and have clarified comorbidity patterns between problematic substance use, behavioural addictions, and mental health disorders [29]. Most relevantly, Guo et al. applied network analysis to problematic use of smartphones and identified central symptoms and key bridge connections to impulsivity and motivational systems, providing a direct methodological precedent for the present study [30]. Collectively, these studies illustrate the unique value of network analysis in revealing complex symptom patterns that may not be captured using conventional approaches. Notably, recent years have witnessed the emergence of network analysis studies specifically examining FoMO and related constructs [31, 32]. The pioneering work by Li et al. employed network analysis to explore relationships among FoMO, smartphone addiction, and social networking site use in Chinese university students [33]. Their study constructed both domain-level and item-level networks, revealing state-FoMO and the "feeling anxious and lost" dimension of smartphone addiction as central nodes. Similarly, Tao et al. applied a two-wave moderation network analysis to examine bidirectional relationships between FoMO, depression, and smartphone addiction [34]. In another longitudinal study, Tao et al. employed symptom network analysis to investigate how prolonged online learning influenced social anxiety and problematic smartphone use among students with different levels of FoMO [35]. Although these studies represent important advances, they primarily examine relationships within a single level of analysis. Li et al.'s item-level network combined items from all three constructs into one unified model [33], whereas Tao et al. examined associations at the dimension level [34, 35]. A key limitation of these approaches is that they do not directly address the theoretical question of how item-level manifestations of a unidimensional construct such as FoMO relate to dimension-level characteristics of a multidimensional construct such as mobile phone addiction. To our knowledge, no previous study has specifically investigated these cross-level relationships using network analysis. The rationale for adopting a mixed-level network approach in the present study rests on both theoretical and methodological considerations. Theoretically, FoMO is conceptualised as a unidimensional cognitive-affective construct [36], and examining it at the item level permits the identification of specific fears that may function as maintaining or bridging mechanisms. Mobile phone addiction, by contrast, is widely regarded as a multidimensional syndrome encompassing distinct facets such as withdrawal, loss of control, and functional impairment [30, 37]. Aggregating these facets into a single total score would obscure potentially differential associations between individual FoMO symptoms and specific dimensions of addictive behaviour. Methodologically, combining items and dimension-level variables within a single network is an established practice in the network psychopathology literature, particularly when the research objective is to examine cross-domain bridge pathways between constructs that differ in their internal structure [30, 34, 38, 39]. This approach balances granularity with parsimony, retaining item-level specificity for the unidimensional construct whilst employing dimension-level aggregation to ensure model stability and interpretability for the multidimensional construct. Guided by these theoretical and methodological considerations, the present study constructs a cross-level network that links individual FoMO items directly to mobile phone addiction dimensions. We calculated centrality and bridge centrality indices to identify central nodes that maintain the overall network and bridge nodes that may facilitate the co-occurrence of these constructs. Based on previous evidence indicating that FoMO is a vulnerability factor for mobile phone addiction, we hypothesised that: (1) the strongest associations would occur within the FoMO and mobile phone addiction communities, respectively; (2) specific FoMO items would exhibit distinct associations with different dimensions of mobile phone addiction; and (3) statistically influential central and bridge nodes would emerge and represent theoretically meaningful targets for understanding the co-occurrence of FoMO and mobile phone addiction. Overall, this study aimed to identify statistically influential nodes that may serve as promising candidates for future intervention research. These nodes are theoretically positioned to contribute to the co-occurrence of FoMO and mobile phone addiction; however, their causal significance requires further validation through longitudinal and experimental studies. Given that no published studies have applied network analysis to investigate the fine-grained associations between FoMO items and dimensions of mobile phone addiction, the present study is both innovative and exploratory in nature. Materials and methods Participants and ethical approval This study employed convenience sampling to recruit college students from three universities in Hunan Province, China. Data were collected through the Questionnaire Star platform (https://www.wjx.cn). Prior to survey administration, three professionally trained psychology researchers introduced the study objectives and procedures after class and emphasized that participation was anonymous and voluntary and that findings would be reported only in aggregate form. Participants were eligible if they were currently enrolled university students who owned a smartphone. Exclusion criteria included: (a) younger than 18 or older than 25 years, or providing invalid age information; (b) responses exhibiting obvious response patterns or regularity (e.g., sequential responses such as "12,345"); and (c) failure to answer attention-check items correctly (e.g., failing to select "disagree" when instructed). A total of 1,690 college students completed the survey. After data screening, 74 cases were excluded because they did not meet eligibility or data quality requirements, resulting in a final analytic sample of 1,616 participants. Participants accessed the questionnaire by scanning a QR code and were free to withdraw at any point. Informed consent was obtained electronically from all participants before beginning the survey. This study was approved by the Ethics Committee of Shanghai Normal University (Approval No. 2025–005) and was conducted in accordance with the Declaration of Helsinki. Measurements Mobile Phone Addiction Index (MPAI) The MPAI, developed by Leung [37] and later revised into Chinese by Huang [40], was used to measure the level of mobile phone addiction tendency among college students. The scale consists of 17 items and can be divided into four subscales, including inability to control craving, such as "Your friends and family have complained about your use of the mobile phone," feeling anxious and lost, such as "When out of range for some time, you become preoccupied with the thought of missing a call," withdrawal and escape, such as "You have used your mobile phone to talk to others when you were feeling isolated," and productivity loss, such as "You find yourself occupied on your mobile phone when you should be doing other things, and it causes problems." Each item is rated on a 5-point Likert scale ranging from 1 (not at all) to 5 (always). The total score ranges from 17 to 85, with higher scores indicating higher levels of mobile phone addiction. A total score of 34 or above is considered indicative of mobile phone addiction. In this study, the MPAI demonstrated excellent internal consistency (Cronbach's α = 0.90). Moreover, the internal consistencies of the inability to control craving dimension, feeling anxious and lost dimension, withdrawal or escape dimension, and productivity loss dimension were also acceptable (α = 0.79, 0.76, 0.72, and 0.70, respectively). Fear of Missing Out scale (FoMOs) The FoMOs, developed by Przybylski et al. [8], is the most widely used instrument for assessing FoMO. A recent meta-analysis [18] reported that 75 of 90 included studies (83.3%) utilized this scale, underscoring its broad acceptance and applicability. In the present study, we adopted the Chinese version revised and validated by Xia et al. [36]. This unidimensional scale consists of 10 items, such as "I fear others have more rewarding experiences than me," rated on a 5-point Likert scale ranging from 1 (not at all true of me) to 5 (extremely true of me). Higher total scores indicate greater levels of FoMO. In the current sample, the internal consistency was excellent (Cronbach's α = 0.89). Data analysis Descriptive statistics were analyzed using SPSS 25.0. Network analysis was conducted in R 4.1.1 using the qgraph and bootnet packages to construct the FoMO and mobile phone addiction network among Chinese college students [41]. The network was estimated via a Gaussian graphical model (GGM) based on nonparametric Spearman correlations [42]. GGM was chosen as it estimates partial correlation networks suitable for continuous and ordinal data [42]. Spearman correlations were used because several variables showed departures from normality. During network construction, we applied the least absolute shrinkage and selection operator (LASSO) regularization combined with the extended Bayesian information criterion (EBIC) to shrink all edges and set negligible ones to zero [43, 44], thus establishing a stable and interpretable network. The EBIC hyperparameter was set to 0.5 to balance sensitivity and specificity in extracting true edges [42], following the default recommendation in the network literature [45]. Within the network model, nodes represented the 10 items of the FoMO scale and the 4 dimensions of the MPAI. Edges between nodes represented partial correlations after statistically controlling for all other nodes [46]. Edge color indicates the direction of the partial correlation: blue edges represent positive correlations, and red edges represent negative correlations. Greater edge saturation indicates a stronger partial correlation. Following established practices in network analysis, a minimum absolute edge weight of 0.03 was considered meaningful for interpretation [47, 48]. The expected influence (EI) of each node was calculated as a centrality index using the R package qgraph [46]. EI is defined as the sum of the edge weights connecting a given node to all other nodes in the network. This metric was selected because it accounts for the sign of edge weights, whereas strength centrality sums absolute values and may yield different results when negative edges are present [49]. A higher EI value signifies more and stronger connections, indicating greater importance within the network structure [49]. Bridge expected influence (BEI) was computed as a bridge centrality indicator using the R package networktools [27]. BEI is calculated as the sum of the edge weights connecting a specific node to all nodes in other pre-defined communities. In our study, the two communities were defined a priori as the FoMO community and the mobile phone addiction community. A higher BEI value indicates stronger connections between communities, reflecting a greater tendency to co-occur with symptoms in the other community [27]. Finally, we employed the bootnet package to assess the accuracy and stability of the network [45]. The accuracy of edge weights was examined through 95% bootstrapped confidence intervals (CIs) based on 1,000 bootstrap samples, with narrower CIs indicating greater precision [45]. The correlation stability (CS) coefficient, calculated using a case-dropping bootstrap approach with 1,000 bootstrap samples, was utilized to evaluate the stability of the EI and BEI estimations, with a value greater than 0.5 indicating strong stability [45]. Bootstrapped difference tests, also employing 1,000 bootstrap samples, were conducted to test the differences in edge weights, EIs, and BEIs. Results For the 1616 participants, the average age was 19.40 ± 1.32 years (M ± SD, range = 18–25 years), and 834 (51.61%) were male. 848 (52.48%) participants reached the threshold of having mobile phone addiction. Table 1 presents the mean scores, standard deviations, EIs, and BEIs for dimensions or items within the FoMO and mobile phone addiction network.Table 1Means, standard deviations, EIs, and BEIs for each variable in the networkVariablesAbbMSDEIBEIFear others' superior experiencesFoMO11.620.860.900.05Fear friends' superior experiencesFoMO21.550.831.010.02Fear of social oblivionFoMO31.700.901.000.11Fear unknown friends' updatesFoMO41.460.800.910.04Need to grasp in-jokeFoMO52.101.050.880.03Wasting time on tracking informationFoMO61.870.921.000.14Bothered missing friend meetupsFoMO71.971.000.880.00Sharing good moments onlineFoMO81.650.910.610.02Bothered missing planned gatheringsFoMO92.141.050.900.07Monitor friends during vacationFoMO101.690.890.930.11Inability to control cravingICC14.674.760.760.06Feeling anxiety and lostFAL7.383.091.030.21Withdrawal and escapeWE6.882.800.710.18Productivity lossPL6.242.630.950.16Abb Abbreviation, M Mean, SD Standard deviation, EI Expected influence, BEI Bridge expected influence Table 1Means, standard deviations, EIs, and BEIs for each variable in the networkVariablesAbbMSDEIBEIFear others' superior experiencesFoMO11.620.860.900.05Fear friends' superior experiencesFoMO21.550.831.010.02Fear of social oblivionFoMO31.700.901.000.11Fear unknown friends' updatesFoMO41.460.800.910.04Need to grasp in-jokeFoMO52.101.050.880.03Wasting time on tracking informationFoMO61.870.921.000.14Bothered missing friend meetupsFoMO71.971.000.880.00Sharing good moments onlineFoMO81.650.910.610.02Bothered missing planned gatheringsFoMO92.141.050.900.07Monitor friends during vacationFoMO101.690.890.930.11Inability to control cravingICC14.674.760.760.06Feeling anxiety and lostFAL7.383.091.030.21Withdrawal and escapeWE6.882.800.710.18Productivity lossPL6.242.630.950.16Abb Abbreviation, M Mean, SD Standard deviation, EI Expected influence, BEI Bridge expected influence Means, standard deviations, EIs, and BEIs for each variable in the network Abb Abbreviation, M Mean, SD Standard deviation, EI Expected influence, BEI Bridge expected influence Network model The FoMO and mobile phone addiction network model is presented in Fig. 1. The network comprised 14 nodes and 63 non-zero edges (weight range = − 0.03 to 0.60). Within the FoMO community, the four strongest edges were between FoMO1 "Fear others' superior experiences" and FoMO2 "Fear friends' superior experiences" (edge weight = 0.60), FoMO7 "Bothered missing friend meetups" and FoMO9 "Bothered missing planned gatherings" (edge weight = 0.37), FoMO3 "Fear of social oblivion" and FoMO4 "Fear unknown friends' updates" (edge weight = 0.30), and FoMO5 "Need to grasp in-joke" and FoMO6 "Wasting time on tracking information" (edge weight = 0.31). Within the mobile phone addiction community, the strongest edges were between ICC "Inability to control craving" and PL "Productivity loss" (edge weight = 0.34), and between ICC and FAL "Feeling anxiety and lost" (edge weight = 0.30).Fig. 1Network structure of FoMO items and mobile phone addiction dimensions. Blue edges represent positive partial correlations, and red edges represent negative partial correlations. The thickness and saturation of the edges are proportional to the strength of the association. All edge weights are provided in Supplementary Table S1. Refer to Table 1 for the node labels and their corresponding descriptions Network structure of FoMO items and mobile phone addiction dimensions. Blue edges represent positive partial correlations, and red edges represent negative partial correlations. The thickness and saturation of the edges are proportional to the strength of the association. All edge weights are provided in Supplementary Table S1. Refer to Table 1 for the node labels and their corresponding descriptions Second, cross-community edges were consistently weaker than within-community connections. Following recommended thresholds, we defined edges with an absolute weight of at least 0.03 as meaningful and focused our interpretation on them. Accordingly, ICC "Inability to control craving" showed no meaningful correlations with the three connected FoMO items (FoMO2, FoMO3, and FoMO6; all weights ≤ 0.02) and can therefore be disregarded. FAL "Feeling anxiety and lost" correlated with six FoMO items (FoMO3, FoMO4, FoMO5, FoMO6, FoMO8, and FoMO9), most strongly with FoMO4 "Fear unknown friends' updates" (edge weight = 0.07). WE "Withdrawal and escape" connected to seven FoMO items (FoMO3, FoMO4, FoMO5, FoMO6, FoMO8, FoMO9, and FoMO10), with the strongest link to FoMO10 "Monitor friends during vacation" (edge weight = 0.11). PL "Productivity loss" was associated with four FoMO items (FoMO1, FoMO3, FoMO6, and FoMO9), most strongly with FoMO6 "Wasting time on tracking information" (edge weight = 0.06). All edge weights in the network are presented in Supplementary Table S1. The 95% CIs of edge weights in the network are narrow, indicating high accuracy of the edge weight estimation (see Supplementary Figure S1). The results of the bootstrapped difference test for edge weights are shown in Supplementary Figure S2. Network centrality The EIs of nodes in the FoMO and mobile phone addiction network are displayed in Fig. 2. The nodes with the highest EI values, and thus the greatest centrality, were "Feeling anxiety and lost" (FAL, EI = 1.03), "Fear friends' superior experiences" (FoMO2, EI = 1.01), "Fear of social oblivion" (FoMO3, EI = 1.00), and "Wasting time on tracking information" (FoMO6, EI = 1.00). The CS coefficient for EI was 0.672, exceeding the recommended threshold of 0.5, indicating good stability of the centrality indices (see Supplementary Figure S3). The result of the bootstrapped difference test for node EI is presented in Supplementary Figure S4.Fig. 2 Centrality indices for each node in the network (raw values). Notably, FAL, FoMO2, FoMO3, and FoMO6 emerged as the nodes with the highest EI values (1.03, 1.01, 1.00, and 1.00, respectively). Refer to Table 1 for the node labels and their corresponding descriptions Centrality indices for each node in the network (raw values). Notably, FAL, FoMO2, FoMO3, and FoMO6 emerged as the nodes with the highest EI values (1.03, 1.01, 1.00, and 1.00, respectively). Refer to Table 1 for the node labels and their corresponding descriptions Network bridge centrality The BEIs of nodes in the FoMO and mobile phone addiction network are displayed in Fig. 3. "Wasting time on tracking information" (FoMO6) and "Feeling anxiety and lost" (FAL) showed the highest BEI values in their respective communities (BEI = 0.14 and 0.21, respectively), identifying them as the key bridge nodes in the FoMO and mobile phone addiction network. The results of the stability test of BEIs are shown in Supplementary Figure S5. The CS coefficient was 0.594, which was greater than 0.5, indicating acceptable stability. The result of the difference test of BEIs is presented in Supplementary Figure S6.Fig. 3 Bridge centrality indices for each node in the network (raw values). Among them, FAL had the highest BEI value (0.21) in the mobile phone addiction community, while FoMO6 had the highest BEI value (0.14) in the FoMO community. Refer to Table 1 for the node labels and their corresponding descriptions Bridge centrality indices for each node in the network (raw values). Among them, FAL had the highest BEI value (0.21) in the mobile phone addiction community, while FoMO6 had the highest BEI value (0.14) in the FoMO community. Refer to Table 1 for the node labels and their corresponding descriptions Discussion This study employed network analysis to investigate the cross-level network structure between FoMO and mobile phone addiction at the symptom level. The strongest edges were observed within each community, along with several meaningful cross-community edges linking FoMO items to mobile phone addiction dimensions. Through centrality and bridge centrality analyses, we identified central and bridge nodes that are statistically prominent within the network. These nodes may represent key elements that are closely associated with the co-occurrence of FoMO and mobile phone addiction, and they warrant further investigation as potential intervention targets. Given the cross-sectional design, however, all such interpretations remain hypothesis-generating; causal or clinical conclusions require future longitudinal and experimental validation. These findings enhance our comprehension of the complex interactive mechanisms between these two constructs, and provide preliminary insights that, if further tested, may inform the development of precise psychological interventions for college students. Correlation pathways between FoMO and mobile phone addiction The results indicated that the strongest edges appeared within communities rather than between them. This pattern is consistent with prior network studies showing that the strongest edges exist within each community when examining relationships between variables in a network consisting of two or more constructs [50–52]. For the FoMO community, some of the strongest positive edges were observed between symptoms capturing fears related to social comparison and between symptoms concerning the monitoring of social activities. This finding aligns with recent longitudinal network analyses, which similarly identified strong intra-community connections among FoMO symptoms [31, 53]. Within the mobile phone addiction community, the most substantial connections involved the emotional dimension of "Feeling anxiety and lost" (FAL) with behavioral components such as "Inability to control craving" (ICC) and "Productivity loss" (PL). Notably, the robust connection between ICC and PL has been consistently identified as one of the strongest edges within the mobile phone addiction network across multiple studies involving Chinese adolescents and university students [33, 54, 55]. This recurring finding aligns with the fundamental premise of network theory that symptoms or sub-components of the same psychological construct naturally exhibit closer interactions and form more tightly coupled clusters [56, 57]. Consequently, it is theoretically expected and empirically supported that the strongest edges would exist within each community. Beyond these within-community connections, we also identified several cross-community pathways. Although these connections were relatively weak, they are theoretically meaningful. Following the recommended threshold for meaningful interpretation [47, 48], we focused our discussion on these interpretable connections. It is noteworthy that "Inability to control craving" (ICC) demonstrated negligible correlations with all FoMO items, falling below this threshold, suggesting that the compulsive craving aspect of addiction may not be directly driven by specific FoMO manifestations. Among the meaningful cross-community links, "Withdrawal and escape" (WE) was most strongly associated with "Monitor friends during vacation" (FoMO10), highlighting the role of digital dependency in non-academic contexts [58]. Within the Chinese collectivistic cultural context, where social integration is strongly emphasized [11], students strive to maintain social proximity during holidays. When unable to participate in real-time activities, they may resort to monitoring friends' online activities to sustain a sense of connection [59]. However, this digital surrogate can precipitate withdrawal symptoms and reinforce avoidance of offline interactions [9, 60, 61], suggesting that real-world social engagement strategies represent a hypothesis that warrants empirical testing in future research. Similarly, "Feeling anxiety and lost" (FAL) was meaningfully connected to several FoMO items, most substantially with "Fear unknown friends' updates" (FoMO4). This aligns with evidence that social anxiety drives problematic phone use [35]. In the Chinese university setting, where social integration is both culturally mandated and closely tied to peer evaluation [10, 11], this anxiety may be especially potent. University students are susceptible to pressures concerning their social standing within the collective and may fear exclusion from social groups if they miss updates [14, 62]. This anxiety triggers feelings of loss and propels frequent phone checking, reinforcing addictive patterns [9, 63]. Whether developing skills to filter socially relevant information could contribute to mitigating this mutual reinforcement remains a question for future investigation. Furthermore, "Productivity loss" (PL) was most strongly linked to "Wasting time on tracking information" (FoMO6), underscoring the detrimental impact of social media on cognitive resources [64]. Excessive engagement can lead to attention deficits and academic decline [65, 66], as students become trapped in "time black holes" of browsing, resulting in procrastination and reduced efficiency [66, 67]. This decline can, in turn, prompt further compensatory phone use [15], creating a self-perpetuating cycle. A meta-analysis has confirmed the link between increased online time and heightened addiction risk [68], suggesting that time management strategies warrant further investigation in longitudinal research. Central symptoms, bridge symptoms, and practical implications Consistent with our hypothesis, influential central and bridge nodes were identified within the FoMO and mobile phone addiction network. We identified four central symptoms based on EI values: "Feeling anxiety and lost" (FAL), "Fear friends' superior experiences" (FoMO2), "Fear of social oblivion" (FoMO3), and "Wasting time on tracking information" (FoMO6), indicating their statistical prominence within the overall network [23]. Furthermore, bridge centrality indices revealed that two of these symptoms, FAL and FoMO6, also served as key bridge nodes, as evidenced by their highest BEI values within their respective communities [22, 27]. This dual role indicates that FAL showed the strongest statistical associations with the mobile phone addiction community, while FoMO6 was most strongly connected to the FoMO community. It must be emphasized that these centrality and bridge centrality indices are derived from a cross-sectional network and therefore reflect statistical associations only; they cannot establish causality, temporal direction, or treatment efficacy. Accordingly, the nodes identified here should be regarded as hypothesis-generating candidates for future research rather than as empirically validated intervention targets. From a network perspective, FAL and FoMO6 are positioned at the intersection of both conditions, lending preliminary statistical support to the hypothesis that they may represent strategically important positions in the co-occurrence pattern. This hypothesis requires direct testing in future experimental and longitudinal research. The network's stability coefficient supports the reliability of these findings. Practical implications With the above cross-sectional constraint clearly acknowledged, the present findings may nonetheless inform hypotheses for future intervention research. Cognitive-Behavioral Therapy (CBT) provides a well-established paradigm that could be tested as a framework for targeting the mechanisms these symptoms may reflect [69, 70]. For the dual-function node FAL ("Feeling anxiety and lost"), which emerged as an emotional bridge, a hypothesis worth testing is whether interventions that address this prevalent negative affective state might help interrupt broader symptom activation [10, 11, 71, 72]. CBT-based emotion regulation modules could be examined for their capacity to help students deconstruct the sequence of environmental triggers, automatic negative thoughts, and compensatory phone use, fostering more balanced evaluations of social information [69, 73, 74]. For the dual-function node FoMO6 ("Wasting time on tracking information"), which emerged as a behavioral bridge, future studies could test whether cognitive restructuring to challenge the dysfunctional belief that compulsive monitoring alleviates FoMO, combined with behavioral experiments and time management training, reduces non-essential tracking [75, 76]. Beyond these targeted approaches, integrating mindfulness principles with CBT represents another testable hypothesis, potentially cultivating non-judgmental awareness of FoMO-related distress and preventing emotional escalation into addictive patterns [75, 77]. For the cognition-focused central symptoms FoMO2 ("Fear friends' superior experiences") and FoMO3 ("Fear of social oblivion"), future studies could evaluate whether interventions guiding students toward healthier perspectives on social comparison and reduced fear of negative evaluation yield clinical benefits, potentially through school-based psychoeducation and social skills training [78, 79]. In summary, the intervention suggestions outlined above represent theoretically derived hypotheses. Their clinical effectiveness and the direction of any therapeutic effects await empirical validation through experimental and longitudinal designs. Limitations and future directions Several limitations should be considered when interpreting the findings of this study. First, the cross-sectional design constitutes a fundamental methodological constraint, as it precludes any inference regarding causal or temporal relationships between network nodes. Furthermore, because the network is estimated at the group level, its structure and central features may not accurately reflect symptom dynamics at the individual level. Critically, bridge centrality in this context reflects statistical connectivity within the network model and must not be interpreted as evidence of temporal or causal bridging between symptoms. To address these limitations, future research should employ longitudinal designs and ecological momentary assessment to track the temporal dynamics of symptom interactions at both the group and individual levels. Second, the use of a unidimensional FoMO scale, while consistent with the most widely adopted measure in the field, presents a specific methodological boundary condition. This choice precluded a dimension-level analysis of the FoMO construct itself, which could have provided additional validation and different perspectives. We recommend that future studies consider employing multidimensional FoMO scales to enable comparative network analyses across different levels. Third, the study relied exclusively on self-reported data, which may be subject to recall bias and social desirability effects. Although the scales demonstrated good reliability, the subjective nature of the data lacks the objectivity afforded by clinical interviews or behavioral measures. Subsequent studies would benefit from integrating multi-method assessment strategies, such as professional diagnostic interviews, experimental tasks, or neurophysiological indices, to triangulate findings. Fourth, the consistently weak cross-community edges observed in our network suggest that the relationship between FoMO and mobile phone addiction operates primarily through indirect psychological mechanisms rather than direct symptom-level pathways. Future studies should therefore investigate potential mediating variables, such as state anxiety or belongingness needs, that may explain these indirect relationships. Additionally, randomized controlled trials are needed to evaluate whether modifying symptoms identified as central in the network produces causal effects on the broader network. Fifth, the network models were estimated without including demographic covariates such as age and gender. While this approach focuses on the intrinsic symptom-level structure, it does not rule out the possibility that these demographic variables may moderate the observed associations. Future studies could examine whether age or gender differences influence the network structure. Finally, the generalizability of our findings is inherently constrained by the specific sample. The participants were exclusively Chinese college students, a population shaped by collectivistic cultural norms, unique social media ecosystems such as WeChat and Douyin, and intense academic pressure. Consequently, the current network structure, including the centrality of specific FoMO items and the bridging roles identified, may be culturally specific and cannot be assumed to generalize beyond this population. Extrapolation to other age groups, cultural settings, or non-student samples should be avoided without direct replication. Future research should validate these findings in more diverse samples, including different age groups, cultural backgrounds, and clinical populations, to establish the broader applicability and boundary conditions of the current network model. Cross-cultural network comparisons, in particular, would help disentangle universal mechanisms from culture-specific expressions of FoMO and mobile phone addiction. Conclusion This study extends prior network research through a fine-grained examination of the cross-level relationships between items of FoMO and dimensions of mobile phone addiction. The network structure revealed stronger connections within each construct and identified several meaningful cross-community pathways. Based on EI values, "Feeling anxiety and lost" (FAL), "Fear friends' superior experiences" (FoMO2), "Fear of social oblivion" (FoMO3), and "Wasting time on tracking information" (FoMO6) were identified as central nodes in the network, highlighting their statistical prominence. Notably, bridge centrality analysis further demonstrated that two of these central symptoms, FAL and FoMO6, also functioned as the most important bridge nodes. These cross-sectional findings are hypothesis-generating in nature; whether the identified nodes represent viable intervention targets requires rigorous causal validation through future longitudinal and experimental research. In summary, this study provides preliminary insights into the fine-grained relationships between FoMO and mobile phone addiction, and offers a foundation for future hypothesis-driven intervention research. Supplementary Information Supplementary Material 1. Supplementary Material 1.