Physical activity linked to lower mobile phone dependence through psychological resources

A cross-sectional survey of 567 college students in northeastern China found that more self-reported physical activity was associated with lower mobile phone dependence, with the overall relationship largely accounted for statistically by psychological capital, self-control and a pathway linking the two. Physical activity had a negative correlation with mobile phone dependence of r = −0.246, p < 0.001, and its total standardized association with dependence was β = −0.255, p < 0.001. Once psychological capital, self-control, gender, age and grade were included, however, the remaining direct association was no longer statistically significant. The results identify possible psychological correlates of the activity–phone-use relationship, but they do not show that exercise causes either psychological resources to increase or mobile phone dependence to fall.

The analysis used PROCESS Model 6 to estimate three indirect associations, with 5,000 bootstrap resamples supplying 95% confidence intervals. Physical activity was indirectly associated with lower mobile phone dependence through psychological capital, with a standardized estimate of −0.069 and a 95% confidence interval from −0.104 to −0.035. A second, larger association ran through self-control, at −0.108, with a 95% confidence interval of −0.157 to −0.060. The sequential pathway—physical activity associated with greater psychological capital, which was in turn associated with greater self-control and lower dependence—had an estimate of −0.036, with a 95% confidence interval from −0.060 to −0.014. None of these intervals included zero.

Psychological capital, abbreviated PsyCap in the study, represents a set of positive psychological resources combining self-efficacy, optimism, hope and resilience. Self-control is a more proximal measure of the ability to regulate thoughts, feelings and actions in pursuit of goals. The researchers placed these constructs in a theoretical sequence in which broader motivational and emotional resources may support behavioral regulation. Their model showed that physical activity was positively associated with psychological capital, β = 0.424, p < 0.001, while both physical activity, β = 0.203, p < 0.001, and psychological capital, β = 0.160, p < 0.001, were associated with higher self-control. Psychological capital and self-control were each inversely associated with mobile phone dependence.

The combined indirect estimate was −0.212, with a bootstrap standard error of 0.030 and a 95% confidence interval of −0.271 to −0.154. Descriptively, this was equivalent to 83.198% of the total activity–dependence association. The individual routes accounted for 26.855% through psychological capital, 42.221% through self-control and 14.122% through the sequential pathway. These percentages describe how the modeled total was distributed among its statistical components; they are not estimates that exercise prevented specific percentages of dependence cases. The self-control route had the largest absolute indirect estimate, although the study did not formally test whether that route was statistically larger than the other two.

Participants were recruited from four comprehensive universities in Northeast China between May and June 2026: Qiqihar University and Jiamusi University in Heilongjiang Province, Beihua University in Jilin Province, and Dalian University in Liaoning Province. The researchers used an online convenience sample recruited through cooperating classes. Of 596 questionnaires collected, 29 were excluded because they were incomplete, completed in implausibly short time or displayed patterned responses, leaving 567 valid cases. Eligibility required regular participation in physical activity for at least one complete academic semester, so the results describe students with some established activity experience rather than the entire range of activity levels among college students.

The sample was not demographically balanced. Sophomores made up 65.96% of the final group, compared with 26.10% freshmen, 4.59% juniors and 3.35% seniors. Men represented 61.02% of participants and women 38.98%; 77.42% were 19 to 22 years old, 21.52% were 18 or younger and 1.06% were 23 or older. The researchers could not report university-specific sample sizes because the survey did not retain university affiliation at the individual level. Class identifiers and the number of participating classes were also removed, preventing an assessment of whether observations were clustered within classes or institutions.

All four main concepts were measured with self-report instruments. Physical activity was assessed with the six-item Activity subscale of the Chinese-language Physical Self-Description Questionnaire. Its items asked about usual activities such as jogging, dancing, swimming and ball games, but did not specify a recall period and did not directly quantify exercise frequency, duration, intensity or energy expenditure. Psychological capital was measured with a 26-item scale covering self-efficacy, optimism, hope and resilience; self-control with the 13-item Brief Self-Control Scale; and mobile phone dependence with five adapted items from the Smartphone Addiction Scale. The authors explicitly interpreted the latter scores as levels of dependence rather than diagnoses of clinical addiction.

The measurement analysis generally supported the questionnaire structure. Cronbach’s alpha was 0.783 for physical activity, 0.923 for psychological capital, 0.881 for self-control and 0.820 for mobile phone dependence. Confirmatory factor analyses showed acceptable-to-good fit for the principal models, and the heterotrait–monotrait ratios, ranging from 0.285 to 0.721, were below the study’s conservative threshold of 0.85. Diagnostics did not suggest that one common factor dominated all 50 responses, although the researchers stressed that such tests cannot rule out common-method variance. Because every construct was assessed in the same survey, social desirability, answer bias and shared response styles remain possible influences.

The study’s design places firm limits on interpretation. Cross-sectional associations cannot establish the temporal order of physical activity, psychological capital, self-control and mobile phone dependence, and the analysis cannot demonstrate that psychological capital caused self-control or that either one caused lower dependence. Convenience sampling, the uneven academic distribution and the requirement for prior physical activity also limit how widely the findings can be generalized. The authors note that future studies should include students with lower activity levels, use objective activity measures and multi-informant or behavioral assessments, retain group identifiers, and examine the relationships longitudinally or experimentally. For now, the evidence supports a nuanced conclusion: in this sample, the negative association between physical activity and mobile phone dependence was statistically consistent with indirect relationships involving broader psychological resources, stronger self-control, and a resource-to-regulation sequence, rather than with a detectable direct relationship after those factors were considered.

Subject of Research: Cross-sectional associations among physical activity, mobile phone dependence, psychological capital and self-control in college students

Article Title: The association between physical activity and mobile phone dependence in college students: serial indirect associations through psychological capital and self-control

Article References: Pan, W., Zhang, Y., & Shan, T. (2026). The association between physical activity and mobile phone dependence in college students: serial indirect associations through psychological capital and self-control. Frontiers in Psychology, 17, Article 1911519. https://doi.org/10.3389/fpsyg.2026.1911519

Image Credits: AI Generated

DOI: 10.3389/fpsyg.2026.1911519

Keywords: physical activity, mobile phone dependence, psychological capital, self-control, college students, cross-sectional survey, indirect associations, bootstrap analysis, China, association, between, physical