Building on the theoretical foundations outlined in Section 2.1, this section focuses on empirical empirical evidence examining the associations between Fear of Missing Out (FoMO) and social media use behaviors, particularly problematic and addictive patterns. Existing studies have consistently shown that individuals with higher levels of FoMO tend to engage more frequently with social media platforms, exhibit compulsive checking behaviors, and experience greater difficulty disengaging from online social interactions (4, 5, 12).

Within digital environments characterized by constant connectivity and real-time social comparison, FoMO has been associated with heightened sensitivity to social cues, such as notifications, updates, and peer activities. These characteristics are closely related to patterns of excessive monitoring and repetitive engagement with social media, which have been identified as core features of social media addiction (2). Empirical findings suggest that FoMO is positively associated with addictive social media use across different populations, including adolescents and university students (5, 13).

Rather than conceptualizing FoMO as a formally tested mediating variable, prior research positions it as a central psychological construct that is closely associated with both emotional insecurity and maladaptive digital engagement. FoMO represents a focal experiential state reflecting individuals’ anxiety about social exclusion and missed opportunities, and has been consistently associated with problematic patterns of social media use. Accordingly, the present study adopts FoMO as a core construct within the proposed framework to examine its association with social media addiction among university students.

While Section 2.1 introduced the core principles of Self-Determination Theory (SDT), this section narrows the focus to empirical studies examining how SDT-related motivational factors are associated with FoMO and digital media behaviors. Research grounded in SDT suggests that variations in autonomous motivation, competence, and relatedness are closely linked to individuals’ psychological experiences in online contexts (10, 14).

Autonomous motivation in social media use has been associated with greater self-regulation and more intentional engagement, whereas controlled or externally driven use is more closely related to individuals’ psychological experiences (11, 13). Individuals who perceive their online engagement as self-endorsed and value-consistent tend to report lower levels of anxiety related to social comparison and online evaluation. Lower levels of such anxiety have been negatively associated with FoMO.

Similarly, perceived competence in managing social media platforms is associated with individuals’ confidence in navigating information flows, regulating attention, and coping with information overload. Empirical evidence indicates that higher levels of digital competence are associated with lower uncertainty and reduced anxiety related to missing social information. These patterns are consistent with a negative association between competence and FoMO-related experiences.

In addition, social support—closely related to the SDT need for relatedness and emphasized within Social and Emotional Learning (SEL) frameworks—has been consistently linked to psychological well-being and adaptive coping. Individuals with stronger offline social support networks tend to report lower dependence on online social validation and reduced vulnerability to FoMO (14). These findings highlight the importance of integrating motivational and social–emotional resources when examining FoMO in digital contexts.

Taken together, prior research suggests that autonomous motivation, competence, and social support are meaningfully associated with FoMO-related experiences in online environments. Rather than proposing causal pathways, the present study builds on these associations to examine how SDT- and SEL-related factors are linked to FoMO and social media addiction within an integrated conceptual framework.

Building on the conceptualization of FoMO as a central psychological construct (see Section 2.2), this study examines the associations between social media–related psychological resources and individuals’ FoMO experiences. As discussed in Section 2.3, autonomous motivation, competence, and social support are associated with individual differences in social media engagement patterns and the interpretation of social information in digital environments.

Individuals with higher levels of autonomous motivation in social media use tend to report more goal-directed and intentional engagement patterns, and lower levels of anxiety related to social comparison and online evaluation. In contrast, lower autonomous motivation has been associated with greater reliance on constant connectivity and higher levels of FoMO.

Similarly, social media competence reflects individuals’ perceived ability to manage information and interactions effectively. Higher perceived competence is associated with more selective engagement and lower uncertainty, whereas lower competence has been associated with greater concerns about informational delay and social exclusion.

Finally, social support is commonly associated with emotional reassurance and a stable sense of belonging, which are linked to lower reliance on continuous online monitoring. When perceived social support is lower, individuals tend to report greater sensitivity to social cues and updates, as well as higher levels of FoMO.

As discussed in Section 2.2, FoMO is conceptualized as a central psychological construct that is closely associated with motivational vulnerabilities and difficulties in emotional regulation within digital environments. Prior research has consistently shown that individuals reporting higher levels of FoMO also report more frequent checking behaviors and higher levels of excessive social media use (4, 5).

Empirical evidence further indicates a strong positive association between FoMO and problematic or addictive patterns of social media use among university students. Rather than implying causal processes, these findings suggest that FoMO is a salient psychological correlate of social media addiction within digitally mediated social contexts. Accordingly, the present study examines the association between FoMO and social media addiction, as specified in the following hypothesis.

In addition to risk-related psychological factors, individual differences may be associated with variations in the relationship between FoMO and social media addiction. Prosocial orientation reflects individuals’ tendencies toward empathy, cooperation, and maintaining interpersonal relationships, which may be related to how they experience and respond to FoMO-related anxiety.

On the one hand, higher prosocial orientation has been associated with greater perceived emotional support and more meaningful social connections, which may be linked to lower levels of distress in social contexts. On the other hand, strong relational sensitivity has also been associated with greater responsiveness to social expectations and perceived obligations in digital contexts. Taken together, these considerations suggest that prosocial orientation may moderate the relationship between FoMO and social media addiction.

Figures 1a,b illustrates the conceptual research framework of this study. Model A represents the hypothesized path model tested using the main structural model, whereas Model B represents the moderation model examined separately using an interaction term approach. Autonomous motivation in social media use, social media competence, and social support are specified as key psychological resources that are hypothesized to be negatively associated with Fear of Missing Out (FoMO). FoMO is conceptualized as a central psychological construct that is positively associated with social media addiction. In addition, prosocial orientation is modeled as a moderating variable that conditions the strength of the association between FoMO and social media addiction. The framework reflects the hypothesized directional associations examined in this study.

The subjects of this study are university students in Taiwan, who are one of the primary groups using social media and are in a critical period of exploring self-identity and establishing social relationships. The specific research subjects include university students from different grades, majors, and backgrounds to ensure the breadth and generalizability of the research results. The study utilized the researcher’s network connections and distributed the electronic questionnaire link through social media groups of university students from different regions and schools between June 1, 2024, and June 30, 2024, to recruit students willing to participate in the study. A total of 422 valid questionnaires were collected, meeting the sample size recommendation of 300–500 by statisticians (15, 16).

The questionnaire for this study is composed of two main sections: basic demographic information and Likert-scale items measuring the core research constructs. The demographic section collects information on participants’ gender, age, and major. The second section includes measurement items related to autonomous motivation in social media use, social media competence, social support, FoMO, social media addiction, and prosocial orientation. To ensure theoretical rigor, the design of each construct were grounded in established literature. The construct of autonomous motivation in social media use was developed based on SDT proposed by Deci and Ryan (9), which emphasizes intrinsic motivation and autonomous motivation as key determinants of self-regulated behavior. The items reflect individuals’ reasons for engaging in social media, such as enjoyment, personal meaning, and value congruence. It should be noted that the items primarily capture autonomous motivation—reflecting intrinsically motivated or value-congruent engagement with social media—rather than autonomous motivation need satisfaction per se. This operationalization is consistent with prior SDT-based studies that focus on motivational regulation rather than basic need fulfillment. Similarly, the measurement of social media competence was derived from SDT’s conceptualization of competence, representing individuals’ perceived effectiveness and mastery in their interactions with digital environments (17–19). The construct of social support was designed with reference to House’s (20) classical social support theory, which highlights the critical role of emotional and instrumental support in promoting psychological well-being. The dimension of FoMO was adapted from Przybylski et al. (3), defining FoMO as the anxiety individuals experience when fearing the loss of meaningful or enjoyable experiences that others may be having. Finally, the construct of prosocial orientation was based on the work of Eisenberg et al. (21, 25), which describes the tendency to engage in behaviors that benefit others—such as empathy, cooperation, and altruism—in social interactions. Overall, the questionnaire integrates validated theoretical frameworks from SDT, social support theory, and prosocial behavior research to comprehensively examine psychological factors associated with social media use and addiction. The full list of measurement items for all constructs is provided in Appendix A for reference. Accordingly, all core constructs in this study are operationalized based on SDT-related motivational and resource-based concepts, while SEL is employed as a complementary interpretive perspective rather than being directly operationalized as competency measures.

This study employs Structural Equation Modeling (SEM) as the primary statistical analysis method to examine the structural relationships among the study variables. SEM integrates Confirmatory Factor Analysis and regression-based path analysis, allowing for the simultaneous examination of multiple relationships among latent constructs. The analysis process consists of two main stages: the Measurement Model and the Structural Model. First, the Measurement Model specifies the relationships between observed indicators and their corresponding latent variables and is evaluated using Confirmatory Factor Analysis (CFA). Second, the Structural Model examines the hypothesized directional associations among the latent variables within the proposed conceptual framework. Finally, a path diagram is used to visually represent the SEM framework, illustrating the specified paths and relationships among variables to facilitate interpretation of the overall model structure.

A total of 422 valid questionnaires were collected. Among the respondents, 219 were male, accounting for 51.9% of the sample. Participants aged 18–20 years constituted the largest age group (n = 224, 53.1%). In terms of academic background, students from medical and healthcare-related majors represented the largest proportion of the sample (n = 145, 34.4%). Detailed demographic information is presented in Table 1.

Following the criteria proposed by Fornell and Larcker (22), convergent validity was assessed using factor loadings, composite reliability (CR), average variance extracted (AVE), and Cronbach’s α. The results show that all factor loadings ranged from 0.750 to 0.917, exceeding the recommended threshold of 0.7. Composite reliability values ranged from 0.872 to 0.925, indicating satisfactory internal consistency. The AVE values ranged from 0.652 to 0.743, all above the 0.5 criterion, and Cronbach’s α coefficients ranged from 0.867 to 0.913. Together, these results indicate that the measurement model demonstrates adequate convergent validity. Detailed results are presented in Table 2.

Discriminant validity was assessed using the Average Variance Extracted (AVE) criterion proposed by Fornell and Larcker (22). The results show that, for most constructs, the square root of the AVE exceeds the correlations with other constructs, indicating adequate discriminant validity. Detailed results are presented in Table 3.

GOF (Goodness of Fit) is an overall measure of the model’s fit, calculated as GOF = AVE ¯ × R 2 ¯ . According to Vinzi et al. (23), a GOF value of 0.1 indicates weak fit, 0.25 indicates moderate fit, and 0.36 indicates strong fit. The result of this study shows a GOF value of 0.507, indicating a strong fit.

The results of the structural model indicate that autonomous motivation in social media use is negatively associated with FoMO (β = −0.138, SD = 0.052, t = 2.655, p = 0.008). Social media competence is also negatively associated with FoMO (β = −0.238, SD = 0.066, t = 3.610, p < 0.001), as is social support (β = −0.202, SD = 0.060, t = 3.390, p = 0.001). In addition, FoMO is positively associated with social media addiction (β = 0.612, SD = 0.034, t = 18.085, p < 0.001).

Beyond statistical significance, the relative magnitudes of the path coefficients provide descriptive information regarding the relative strength of these associations. Among the three antecedents of FoMO, social media competence shows the strongest negative association (β = −0.238), followed by social support (β = −0.202), whereas autonomous motivation in social media use exhibits a comparatively weaker, though still significant, negative association (β = −0.138). This pattern indicates that students’ perceived ability to effectively manage social media environments shows a stronger negative association with FoMO than motivational autonomy alone.

Furthermore, the strong positive association between FoMO and social media addiction indicates that higher levels of FoMO co-occur with higher levels of excessive and compulsive social media use. Taken together, these findings highlight the potential relevance of digital competence and social support in understanding FoMO and its association with problematic social media use. Table 4 presents the detailed results of the path analysis, and Figure 2 illustrates the structural model.

SmartPLS provides established procedures for testing moderation effects through the construction of interaction terms (24). Following this approach, moderation was examined using a separate analysis in which an interaction term between FoMO and prosocial orientation was specified. Consistent with standard practice, this moderation analysis was conducted independently from the main structural model, which focuses on the hypothesized direct associations among the core constructs.

The results show that the interaction term between FoMO and prosocial orientation is significantly associated with social media addiction, indicating a statistically significant moderation effect. For clarity of presentation, the moderation effect is reported separately from the main path model. Detailed results of the moderation analysis are presented in Table 5, and the interaction pattern is illustrated in Figure 3.

This study makes several theoretical contributions to the literature on social media addiction and digital well-being. First, it conceptually integrates Self-Determination Theory (SDT) and Social and Emotional Learning (SEL) to provide a more comprehensive framework for understanding psychological correlates of FoMO and social media addiction. By jointly considering motivational needs and social–emotional competencies, the study extends existing research that has often examined these perspectives in isolation.

Second, this study positions FoMO as a central psychological construct within an integrated SDT–SEL framework. Rather than conceptualizing FoMO as a formally tested mediating variable, the findings highlight its role as a focal psychological factor that is closely associated with motivational need–related experiences and addictive patterns of social media use. This conceptual positioning helps clarify FoMO’s importance in linking motivational and emotional processes to digital behaviors without overstating causal or mediational claims.

Third, the study contributes to the literature by identifying prosocial orientation as a contextual moderator that strengthens the association between FoMO and social media addiction. This finding challenges the predominantly one-dimensional view of prosocial orientation as uniformly protective and underscores its complex, context-dependent role in digitally mediated social interactions. By showing that prosocial traits may be associated with greater vulnerability to addictive behaviors under certain conditions, the study opens new avenues for theoretical exploration of social motivation in digital environments.

The findings of this study offer several practical implications for educational institutions, mental health practitioners, and policy makers concerned with promoting healthy digital engagement among university students. First, the negative associations between autonomous motivation, competence, social support, and FoMO suggest that initiatives aimed at enhancing students’ self-regulation, digital literacy, and offline social resources may be beneficial. Educational programs that emphasize reflective social media use, goal-aligned engagement, and information management skills could support healthier patterns of use.

Second, given the positive association between FoMO and social media addiction, early identification of FoMO-related anxiety may be relevant for informing prevention-oriented support efforts. Counseling services and student support systems may consider incorporating FoMO-related assessments and interventions that focus on emotional regulation, self-management, and coping with social comparison.

Finally, the moderating role of prosocial orientation underscores the relevance of balanced approaches when encouraging social engagement and altruistic behaviors. While fostering empathy and cooperation remains valuable, practitioners should also emphasize boundary setting, digital self-care, and mindful connectivity. Encouraging students to maintain healthy limits in online responsiveness may be helpful in reducing the likelihood that prosocial tendencies become associated with excessive or compulsive social media use.

Despite the contributions of this study, several sampling-related limitations should be acknowledged. First, although the sample consists of university students in Taiwan, a notable proportion of the respondents (approximately 34.4%) are from medical and healthcare-related majors. Students in these disciplines often experience relatively high academic pressure, frequent information exposure, and intensive peer coordination through digital platforms. As a result, their patterns of social media use and psychological responses—particularly regarding Fear of Missing Out (FoMO)—may differ from those of students in other academic fields. This disciplinary concentration may therefore introduce a potential disciplinary bias and limit the generalizability of the findings to students in less academically intensive or differently structured programs.

Before discussing general methodological limitations, we first address sampling-specific considerations that are directly relevant to the interpretation of the present findings. Despite the contributions of this study, several limitations related to sampling characteristics should be acknowledged. First, a notable proportion of the sample (approximately 34%) consisted of students majoring in medical and healthcare-related fields. Students in these disciplines may exhibit higher levels of academic pressure, information sensitivity, and reliance on digital communication for academic coordination and peer support. As a result, their patterns of social media use and psychological responses to online information may differ from those of students in other academic fields. Although medical and healthcare majors represent an important segment of the university population, this disciplinary concentration may limit the generalizability of the findings to students in less academically intensive or differently structured programs.

Second, the recruitment strategy relied on social media groups targeting university students, which may introduce a degree of self-selection bias. Students who are more active on social media platforms are more likely to encounter and respond to online surveys, potentially leading to an overrepresentation of frequent social media users. Consequently, the observed levels of FoMO and social media addiction may be higher than those found in the general university student population. This sampling characteristic suggests that the results should be interpreted as reflecting psychological mechanisms among relatively active social media users, rather than serving as prevalence estimates applicable to all students.

Importantly, these sampling characteristics do not undermine the internal validity of the proposed theoretical relationships. The primary objective of this study was to examine the associative mechanisms linking social media–related psychological resources, FoMO, and social media addiction, rather than to estimate population-level prevalence. Nevertheless, future research could adopt stratified or multi-stage sampling strategies, recruit participants through offline channels, or conduct cross-disciplinary comparisons to enhance representativeness and further validate the robustness of the proposed model.

Despite its contributions, this study has several limitations that warrant consideration. First, the sample was limited to university students in Taiwan, which may constrain the cross-cultural generalizability of the findings. Future research could conduct cross-cultural comparative studies to examine whether the observed associations hold across different cultural and social contexts. Second, the research model primarily focused on autonomous motivation, competence, social support, FoMO, and prosocial orientation, while other potential influencing factors—such as personality traits, family background, and academic stress—were not included. Future studies could incorporate these variables to provide a more comprehensive understanding of factors associated with social media addiction. Third, this study adopted a cross-sectional design, which limits the ability to establish causal relationships among variables. Future research could employ longitudinal or experimental designs to further examine the temporal ordering and dynamic relationships among the constructs. Finally, as this study relied on self-reported questionnaires, there may be risks of common method bias and social desirability effects. To enhance validity, future studies are encouraged to adopt multi-method approaches, such as behavioral data analysis, digital trace tracking, or qualitative interviews, to triangulate findings and strengthen the robustness of the conclusions.