Academic Stress Fuels Gaming Addiction Through Anxiety, Peer Support Buffers It

Academic stress is a known risk factor for internet gaming addiction among college students, but the underlying mechanisms have remained unclear. A new study published in Frontiers in Psychiatry examined how anxiety mediates the relationship between academic stress and internet gaming addiction, and whether peer support moderates this pathway. The research offers a detailed psychological model of how stress translates into problematic gaming behavior.

The study surveyed 636 Chinese college students using a cross-sectional design. Participants completed validated measures of academic stress, anxiety, peer support, and internet gaming addiction. Researchers used the PROCESS macro for SPSS, applying Model 4 to test mediation and Model 59 to test moderated mediation, controlling for age and gender across all analyses.

Results showed that academic stress was positively associated with internet gaming addiction (r = 0.28, p < 0.001). Anxiety significantly mediated this relationship, with an indirect effect of 0.22 (95% CI [0.16, 0.27]). The total effect of academic stress on gaming addiction was 0.382, with the indirect pathway through anxiety accounting for 56.3% of that total effect, indicating partial mediation.

Peer support significantly moderated the indirect effect by buffering the path from academic stress to anxiety (b = -0.18, p < .01). For students with lower peer support, the relationship between academic stress and anxiety was stronger (b = 0.42, p < .001). For those with higher peer support, the relationship remained significant but was considerably weaker (b = 0.24, p < .001).

Notably, peer support did not significantly moderate the direct path from academic stress to gaming addiction, nor the second stage of the mediating pathway from anxiety to gaming addiction. This suggests the buffering effect of peer support is stage-specific: it weakens the stress-to-anxiety link but does not alter how anxiety drives gaming behavior once aroused.

The researchers integrated three theoretical frameworks to explain the findings. General strain theory explains why academic stress elevates addiction risk by triggering negative emotions. Stress and coping theory clarifies how anxiety transforms stress into maladaptive coping behaviors such as immersive gaming. Ecological systems theory identifies the social context—peer support—as the boundary condition under which this process is attenuated.

The study’s practical implications are significant. Schools can strengthen student development support systems to alleviate academic stress, offer emotional management courses to enhance emotion regulation skills, and promote peer support through organized peer groups and tutoring programs. Strengths-based cognitive behavioral therapy may also be provided to alleviate anxiety before it escalates into problematic gaming.

Several limitations warrant caution. The cross-sectional design cannot establish temporal precedence or causality, and reverse causation is plausible—problematic gaming may itself lead to academic difficulties that increase stress and anxiety. The sample was drawn solely from Chinese college students at one university using convenience sampling, raising questions about generalizability. Self-report measures are also susceptible to social desirability bias.

Despite these limitations, the findings highlight the importance of anxiety management and peer support interventions for preventing internet gaming addiction among college students facing academic pressure. The stage-specific buffering effect of peer support offers a precise target for intervention: strengthening peer relationships may reduce the anxiety that transforms academic stress into gaming addiction, even if it cannot fully eliminate the direct effects of stress on gaming behavior.

The statistical architecture of the study deserves closer attention, because the distinction between mediation and moderation carries real consequences for how the findings should be interpreted. Mediation, tested here with PROCESS Model 4, addresses the question of how academic stress relates to gaming addiction: the stress signal travels through anxiety, which then fuels gaming behavior. Moderation, tested with Model 59, addresses a different question—under what conditions that pathway strengthens or weakens. By combining the two in a single moderated mediation framework, the researchers were able to pinpoint exactly where in the causal chain peer support exerts its influence, rather than simply asking whether peer support matters in some general sense. The answer they obtained—that peer support buffers only the first stage, the transition from stress to anxiety—narrows the intervention window considerably and gives the findings a precision that broad correlational studies often lack.

The magnitude of the reported associations is worth contextualizing. A correlation of 0.28 between academic stress and internet gaming addiction is conventionally described as a small-to-moderate effect, meaning that academic stress explains roughly eight percent of the variance in gaming addiction scores on its own. This is typical for psychological research on behavioral addictions, where outcomes are shaped by many interacting factors—personality traits, family environment, sleep, academic performance, and access to gaming, among others. The indirect effect through anxiety, with a confidence interval comfortably excluding zero, indicates that the mediating pathway is statistically reliable, but the partial nature of the mediation means that other routes from stress to gaming remain unexplored. Possible candidates suggested by prior literature include escape and coping motives, depressive symptoms, loneliness, and self-esteem, each of which could carry part of the remaining effect.

The choice of anxiety rather than depression as the mediator is theoretically meaningful. Anxiety is characterized by heightened arousal, worry, and an urge to escape aversive internal states, which aligns closely with the appeal of immersive games that offer immediate absorption and distraction. Depression, by contrast, is often marked by anhedonia and withdrawal, which may relate to different patterns of media use. The stress and coping framework invoked by the authors predicts precisely this: when students appraise academic demands as exceeding their coping resources, the resulting anxious arousal motivates avoidant, emotion-focused strategies. Gaming provides rapid, reliable relief from anxious states, and this reinforcement loop—stress, anxiety, escape, temporary relief—can gradually consolidate into the pattern of persistent, compulsive use that defines internet gaming addiction.

The buffering role of peer support fits within a long tradition of research on social buffering of stress. Supportive relationships are thought to dampen physiological and emotional stress responses through several mechanisms: they provide emotional reassurance that reframes threats as manageable, they offer instrumental help such as study assistance that directly reduces academic strain, and they occupy time and attention that might otherwise flow toward solitary gaming. The finding that high peer support reduced the stress-to-anxiety coefficient from 0.42 to 0.24—while leaving the anxiety-to-gaming link intact—suggests that peers act upstream, at the point of emotional generation, rather than downstream at the point of coping choice. In practical terms, a well-supported student may still experience the pull of gaming when anxious, but is less likely to become anxious in the first place.

The epidemiological backdrop in China gives the study added relevance. With internet game users numbering 573 million as of December 2025 according to national statistics, and an estimated prevalence of internet gaming addiction near 7.5 percent among Chinese college students, even small shifts in population-level risk factors could translate into substantial public health impact. College represents a developmental window in which academic evaluation intensifies, parental oversight recedes, and peer networks become the dominant social context—conditions that simultaneously elevate academic stress and expand the influence of peers, for better or worse. The ecological systems perspective underlying the moderation hypothesis captures this: the microsystem surrounding a student determines how a given stressor is experienced and metabolized.

Several methodological features shape how confidently the results can be generalized. The moderated mediation model was estimated with age and gender controlled, which addresses basic demographic confounding but leaves open the possibility that unmeasured variables—such as trait impulsivity, family functioning, or prior gaming history—account for part of the observed associations. The reliance on self-report instruments means that all four constructs were measured through subjective ratings, and common-method variance can inflate correlations between them. The single-university convenience sample, while adequate for testing the hypothesized model, limits extrapolation to students in other regions, institutions of different selectivity, or non-Chinese cultural contexts where norms around gaming and help-seeking may differ substantially.

The cross-sectional design imposes the most important interpretive constraint. Although the model is drawn as a causal sequence—stress to anxiety to gaming—the data were collected at a single time point, so the direction of the arrows reflects theoretical assumption rather than demonstrated temporal order. Longitudinal designs, in which stress is measured before anxiety and anxiety before subsequent gaming behavior, would provide a stronger test. Experimental or intervention studies would go further: if programs that reduce anxiety or strengthen peer networks produce measurable declines in problematic gaming, the causal claims would gain considerable support. Such studies would also clarify whether the buffering effect of peer support can be deliberately cultivated, for instance through structured peer mentoring, cooperative learning arrangements, or residence-based community building.

For clinicians and campus mental health services, the stage-specific nature of the buffering effect suggests a two-pronged approach. First, screening for anxiety among students reporting high academic stress could identify those on the pathway toward problematic gaming before compulsive patterns consolidate. Second, interventions that build peer connection—rather than targeting gaming behavior directly—may prevent the anxious states that motivate escape in the first place. The finding that peer support does not weaken the anxiety-to-gaming link cautions against expecting social connection alone to remediate established addiction; once anxiety-driven gaming is entrenched, more direct clinical treatment may be needed. Future research replicating the model in longitudinal and multi-site samples, and testing additional mediators and moderators, will determine how robust and transportable these findings are across the diverse populations of students navigating academic pressure worldwide.

Subject of Research: Psychology

Article Title: Academic stress and internet gaming addiction in college students: a moderated mediation model with anxiety as mediator and peer support as moderator

Article References: Li, J., Liu, B., & Wang, Y. (2026). Academic stress and internet gaming addiction in college students: a moderated mediation model with anxiety as mediator and peer support as moderator. Frontiers in Psychiatry, 17, Article 1958057. https://doi.org/10.3389/fpsyt.2026.1958057

Image Credits: AI Generated

DOI: 10.3389/fpsyt.2026.1958057

Keywords: academic stress, anxiety, internet gaming addiction, college students, peer support, moderated mediation, mental health, China, Academic, stress, internet, gaming