Introduction

Sleep quality is a critical factor influencing both physiological and psychological health. Insufficient or poor-quality sleep can lead to a range of adverse outcomes, including reduced concentration, heightened negative emotions such as anxiety, and increased levels of depression. Poor sleep quality is highly prevalent among college students, characterized by insufficient sleep duration and a significant decline in sleep quality1. Alarmingly, irregular sleep–wake patterns and inadequate sleep time are widespread in this population2. Poor sleep quality not only directly contributes to mental fatigue and impaired learning ability but also leads to mood fluctuations and cognitive decline, further exacerbating mental health issues3. Therefore, exploring the factors that affect sleep quality and identifying effective strategies to improve sleep quality among college students holds substantial practical significance.

Physical activity and sleep quality

Physical activity is defined as bodily movement produced by skeletal muscle contraction, leading to a substantial increase in energy expenditure4. It is a complex, multi-dimensional behavior that can be further categorized based on its frequency, duration, and intensity4. Research demonstrates significant associations between physical activity and multiple dimensions of physical function and psychological well-being5. As a non-pharmacological intervention, physical activity improves sleep quality by alleviating anxiety and depression, common contributors to sleep disturbances6. Research indicates that being physically active is strongly associated with improved sleep quality among college students7. One systematic review and meta-analysis revealed that engaging in moderate-to-high intensity physical activity is linked to better sleep outcomes among college students8. Moreover, there is an interactive relationship between high levels of physical activity and good sleep quality, suggesting that increased activity promotes more restorative sleep9. In contrast, physical inactivity is associated with poor sleep quality7. Low levels of physical activity are independently and interactively linked to an elevated risk of sleep disturbances among college students10. In addition, sedentary behavior has been found to have a significant impact on poor sleep quality, further underscoring the importance of regular physical activity11. In summary, evidence supports a clear positive relationship between physical activity and good sleep quality, with higher activity levels promoting better sleep quality, and sedentary behavior or inactivity contributing to poorer sleep quality. These findings highlight physical activity as a key strategy for improving sleep quality among college students. Therefore, Hypothesis 1 is proposed: Physical activity has a predictive effect on sleep quality among college students.

The mediating role of rumination

Rumination is a negative cognitive pattern and coping style formed by individuals under the influence of long-term negative events and is a cognitive susceptibility factor leading to negative emotions12. The Stress-Buffering Hypothesis proposes that protective factors can mitigate the negative impacts of stress on health through attenuating stress appraisal and interrupting the pathway between stress and health deterioration13. Physical activity is one of the protective factors, while rumination serves as a pivotal mediator in the pathway between stress and health outcomes14. Empirical research finds that long-term physical exercise can reduce rumination15. Also, moderate-intensity aerobic exercise can reduce rumination16. It has been found that physical activity can significantly predict a negative correlation with rumination in college students17. Rumination is closely related to sleep quality. Rumination keeps the individual focused on the negative content of the event and the negative impact it has, perpetuating negative emotional experiences such as feelings of loneliness and depression, which can lead to adverse health consequences such as poor sleep quality18. Consistent with the Stress-Buffering Hypothesis, rumination represents a maladaptive cognitive process that is activated by stressful experiences and subsequently impairs sleep quality through prolonged cognitive hyperarousal19. The higher the level of rumination, the poorer sleep quality after an individual experiences an acute stressful event20. Rumination is linked to poor sleep quality in young adults21. The higher the level of rumination, the worse the sleep quality of college students22. Further, rumination has associations with short- and long-term sleep disturbances among college students23. Therefore, this study proposes Hypothesis 2: Rumination plays a mediating role in the relationship between physical activity and sleep quality among college students.

The mediating role of psychological resilience

Psychological resilience is the ability to help individuals recover well from traumatic events and stressful situations24. Physical activity has an important effect on psychological resilience among college students25. Physical activity is positively associated with college students’ psychological resilience25. The research found that physical activity was significantly positively correlated with the psychological resilience of college students to varying degrees, and the psychological resilience of college students was gradually improved with the improvement of physical activity level26. Studies revealed that endurance-enhancing physical activity interventions can increase psychological resilience27. As a positive psychological trait, psychological resilience has a significant impact on sleep quality28. According to the Stress-Buffering Hypothesis, psychological resilience serves as an internal protective factor that enables individuals to mitigate perceived stress and adopt more adaptive coping strategies, thereby reducing the likelihood of sleep quality impairments13. One study reported that those who self-reported poor sleep quality obtained lower resilience scores than those reporting good sleep quality, and participants with poorer sleep quality were at higher risk of low resilience relative to those with good sleep quality29. Those with shorter duration sleep were also at higher risk for low psychological resilience, relative to those with longer duration sleep29. Further, the positive correlation between poor sleep quality and depressive symptoms was strongest in low-level resilience students30. The increased psychological resilience also weakened the relationships between perceived stress and anxiety on sleep quality31. Physical activity generates positive emotions that dismantle rumination, ultimately improving sleep quality32. Therefore, Hypothesis 3 is proposed: Psychological resilience plays a mediating role in the relationship between physical activity and sleep quality among college students.

The chain mediating role of rumination and psychological resilience

Rumination is considered to be a pathological way of thinking in which an individual, after experiencing a negative life event, repeats thoughts about the causes and potential consequences of the negative event without thinking about how to improve or solve the problem. Psychological resilience can help individuals move on from negative emotions and return to normal life as quickly as possible33. Rumination can affect an individual’s level of psychological resilience34. Research reveals that rumination is negatively correlated with psychological resilience, the higher the level of rumination, the lower the psychological resilience35. Higher psychological resilience prevents ruminative thinking from exacerbating or worsening the individual’s experience of depressed mood26. Rumination can lead to disruption of an individual’s cognitive patterns, resulting in a decrease in psychological resilience, which in turn affects sleep quality36. According to the Stress-Buffering Hypothesis, psychological resilience acts as a personal resource that buffers the negative effects of stress on physical and mental health, such as sleep quality13. Research indicates that psychological resilience serves as a moderating factor in the relationship between rumination and sleep quality29. Physical activity reduces psychological stress by reducing rumination, a stress-related cognitive pattern, while psychological resilience further alleviates the negative effects of psychological stress on sleep quality. Based on this, Hypothesis 4 is proposed: Rumination and psychological resilience serially mediate the relationship between physical activity and sleep quality.

In summary, the interrelationships between physical activity, sleep quality, rumination, and psychological resilience in college students have garnered increasing research attention. Existing studies have primarily focused on the direct association between physical activity and sleep quality37, the independent mediating role of rumination in this relationship38,39, and the distinct mediating effect of psychological resilience26. However, the combined mediating effects of rumination and psychological resilience in the relationship between physical activity and sleep quality remain unexplored. Rumination is typically associated with negative emotions and sleep disturbances, while psychological resilience serves as a positive psychological resource. This study combined rumination with psychological resilience to explore the relationship between physical activity and sleep quality, providing a new perspective to reveal the psychological mechanism. The proposed sequential mediation model is shown in Fig. 1.

Fig. 1
figure 1

Theoretical framework.

Full size image

Materials and methods

Participants and procedure

The convenience sampling method was used to investigate college students in December 2023. The participants came from five universities in Guangdong province, covering professional fields such as humanities and social sciences, engineering technology, medicine and life sciences, economics and management, agriculture, and so on. Questionnaire Star was used to conduct an online survey and spread it on campus social platforms such as WeChat groups, QQ groups, and campus forums. To improve the representativeness and quality of the data, this survey is password-protected, and the password is provided to participants from five universities. A total of 1250 questionnaires were collected. After excluding 96 invalid responses (13 due to short completion time, 31 due to missing data, 16 due to response patterns, and 36 due to inconsistent responding), 1154 valid questionnaires were retained, resulting in an effective response rate of 92.3%. The study protocol was approved by the Academic Ethics Committee of Zhaoqing University. All participants voluntarily took part in the study, provided informed consent, and committed to answering all questions truthfully. The total scores of the responses were used to determine the scoring of all the surveys. Items in each scale are in a fixed order.

Research tools

Physical activity rating scale (PARS-3)

The Physical Activity Rating Scale (PARS-3) was developed by Japanese scholar Kimio Hashimoto and compiled by Liang40. It is used to assess an individual’s level of physical activity. The scale comprises three items: intensity, duration, and frequency. Each item is measured using a 5-point scale from 1 to 5. The total score is calculated as follows: intensity × (duration − 1) × frequency. Higher scores indicate higher levels of physical activity. In the previous study, this measurement tool has been widely verified and applied in Chinese college students41,42.

Ruminative responses scale (RRS)

The Ruminative Responses Scale (RRS) was revised by Han and Yang based on Nolen-Hoeksema43. It is used to assess an individual’s rumination conditions, describing a focus on the self, symptoms, causes, and consequences of depression. The scale includes three dimensions and 22 items: symptomatic rumination (12 items), compulsive thinking (5 items), and reflective thinking (5 items). Each item was scored on a 4-point Likert scale ranging from 1 (never) to 4 (always). Higher scores indicate higher levels of rumination. This measurement tool has been widely used with Chinese college students and has demonstrated good reliability44,45.

Brief resilience scale (BRS)

The Brief Resilience Scale (BRS), revised by Chen et al., was used to assess the psychological resilience of college students46. The scale comprises three positively keyed and three reverse-scored items, assessing an individual’s psychological resilience and adaptive capacity when confronting stressful or adverse situations. A 5-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree) was used. Higher scores indicate a higher level of psychological resilience. This measurement tool has good validity in a population of Chinese college students26,47.

Pittsburgh sleep quality index (PSQI)

The Pittsburgh Sleep Quality Index Scale (PSQI) was developed by Buysse and revised by Liu et al48. It is used to assess an individual’s sleep quality over the past month. The scale includes 18 questions with 7 dimensions: sleep quality, time to sleep, sleep time, sleep efficiency, sleep disorders, hypnotics, and daytime dysfunction. Each component is scored on a scale from 0 to 3, resulting in a total score range of 0 to 21. The higher the score, the poorer the sleep quality. This measurement tool has been extensively validated and widely utilized in Chinese college students37,49.

Statistical analysis

Data analysis was conducted using SPSS 25 and AMOS 24, employing both descriptive statistics and structural equation modeling (SEM). First, following data screening (addressing missing values, outliers, and normality), we conducted descriptive and correlational analyses of demographic variables and study constructs. Then, according to Anderson and Gerbing50, we first examined the measurement model’s reliability and validity before testing the structural relationships. Finally, a serial mediation model was analyzed to assess both direct and indirect effects of physical activity on sleep quality, thereby testing our research hypotheses. Parameter estimates for the measurement and structural models were performed using the maximum likelihood method in the SEM analysis. Model fit was evaluated using multiple indices, including chi-square (χ2), normed chi-square (χ2/df), comparative fit index (CFI), Tucker–Lewis index (TLI), root mean square error of approximation (RMSEA), and standardized root mean square residual (SRMR), with acceptable thresholds set at χ2/df < 3, CFI/TLI ≥ 0.95, RMSEA < 0.08, and SRMR < 0.0551,52,53. Bootstrapping with 5000 resamples was implemented to test the mediation effects54.

Results

The study included 1154 participants (648 males, 56.15%; 506 females, 43.85%) with a mean age of 20.60 years (SD = 1.31). The sample comprised 339 freshmen (29.38%), 305 sophomores (26.43%), 286 juniors (24.78%), and 224 seniors (19.41%) (see Table 1).

Table 1 Characteristics of the participants.
Full size table

Before SEM analysis, data screening procedures were conducted to verify assumptions, including sample size adequacy (N = 1154), outlier detection, and normality assessment. The Mahalanobis D2 measure showed no potential multivariate outliers in this study52. Normality assessment results (Table 2) indicated acceptable distributions, as evidenced by skewness values ranging from − 0.400 to 0.857 and kurtosis values from − 0.355 to 0.664, satisfying established criteria53.

Table 2 Descriptive analysis, normality, correlation, reliability, and discriminant validity.
Full size table

Next, the samples were characterized, and bivariate correlation analyses were performed. As shown in Table 2, the average score of physical activity was 27.384 (SD = 26.035), the average score of rumination was 44.710 (SD = 14.659), the average score of psychological resilience was 18.798 (SD = 5.377), and the average score of sleep quality was 5.696 (SD = 2.846). Additionally, all bivariate correlation estimates were statistically significant (p < 0.01). Physical activity negatively correlated with rumination and sleep quality (r =  − 0.301, p < 0.01; r =  − 0.258, p < 0.01), and positively correlated with psychological resilience (r = 0.242, p < 0.01). Rumination is negatively correlated with psychological resilience (r =  − 0.286, p < 0.01) and positively correlated with sleep quality (r = 0.359, p < 0.01). Psychological resilience is negatively correlated with sleep quality (r =  − 0.430, p < 0.01).

According to Anderson and Gerbing50, we conducted CFA on the four measurement models and the overall measurement model before structural model analysis. Table 3 shows that all standardized factor loadings were above 0.7 and statistically significant (p < 0.001), justifying the retention of all items53. The overall measurement model demonstrated good fit to the data, with χ2 = 108.134, p < 0.001, χ2/df = 2.703, CFI = 0.978, TLI = 0.970, CFI/TLI = 1.008, RMSEA = 0.054, SRMR = 0.027. All constructs demonstrated acceptable convergent validity, with AVE values of 0.704 and 0.604, and item reliabilities ranging from 0.531 to 0.781, collectively satisfying the 0.5 threshold. Composite reliabilities were 0.877 and 0.901, all exceeding the 0.7 standard. Collectively, these results provide strong evidence supporting the convergent validity of the measurement models.

Table 3 Confirmatory factor analysis (CFA) results for the measurement model.
Full size table

As shown in Fig. 2, the structural model analysis examined the path coefficients between variables and the proportion of variance explained by exogenous variables. The results indicated that physical activity significantly negatively predicts sleep quality (β =  − 0.090, p < 0.001). Further, physical activity significantly negatively predicts rumination (β =  − 0.331, p < 0.001), and rumination positively predicts sleep quality (β = 0.242, p < 0.001). Besides, physical activity significantly positively predicts psychological resilience (β = 0.158, p < 0.001), and psychological resilience significantly negatively predicts sleep quality (β =  − 0.346, p < 0.001). Rumination also negatively predicts psychological resilience (β =  − 0.293, p < 0.001). Moreover, the squared multiple correlation (R2) for sleep quality was 0.275, indicating that physical activity, rumination, and psychological resilience together explained 27.5% of the variance in sleep quality, demonstrating that the proposed research model is acceptable.

Fig. 2
figure 2

Results of structural model analysis with standardized coefficients and R2. Note PA, physical exercise; PR, psychological resilience; SQ, sleep quality. The path coefficients in the figure are standardized estimates.

Full size image

Finally, to validate the proposed hypotheses, this study evaluated the direct and indirect effects of physical activity on sleep quality. As shown in Table 4, physical activity had significant direct (β =  − 0.010, Z =  − 2.500, p < 0.01) and indirect effects (β =  − 0.018, Z =  − 6.000, p < 0.001) on sleep quality, indicating that the structural model is partially mediated. Further specific indirect effects indicated that physical activity negatively predicted sleep quality through rumination (β =  − 0.009, Z =  − 4.500, p < 0.001) or psychological resilience (β =  − 0.006, Z =  − 3.000, p < 0.01). Meanwhile, rumination and psychological resilience sequentially mediated the relationship between physical activity and sleep quality (β =  − 0.004, Z =  − 4.000, p < 0.001). Moreover, the results from 5,000 bootstrap samples showed that their confidence intervals did not contain zero, providing further evidence for the direct and indirect effects of physical activity on sleep quality. Additionally, this study compared these three specific indirect effects and found that the sequential PA-Rumination-PR-SQ pathway had greater explanatory importance than the two simple mediation paths (see Table 4).

Table 4 Direct, indirect, and total effects of the statistical model.
Full size table

Discussion

In summary, the interrelationships between physical activity, sleep quality, rumination, and psychological resilience in college students have garnered significant research attention. While existing studies have established the direct link between physical activity and sleep quality37, the independent mediating roles of rumination38,39 and psychological resilience26 in this relationship, a critical gap remains. Specifically, the potential serial mediating pathway involving both rumination (a maladaptive cognitive process) and psychological resilience (an adaptive resource) between physical activity and sleep quality has not been empirically investigated. This study addresses this gap by proposing and testing a novel conceptual model (Fig. 1) that posits rumination and psychological resilience act as sequential mediators.

The direct effect of physical activity on sleep quality

The present study showed that physical activity negatively predicted sleep quality among college students. This result supports Hypothesis 1 and is consistent with previous findings in college student populations7,37. As modern health concepts continue to evolve and advance55,56, the beneficial effects of physical activity on individuals’ physical and psychological well-being have become increasingly evident. Physical activity is a critical factor influencing sleep quality57. Research indicates that moderate exercise is more effective than vigorous or acute exercise in enhancing sleep quality6,58. The study also found that moderate-intensity physical activity positively impacts sleep quality, whereas high-intensity physical activity shows no significant effect59. Regular physical activity yields small but beneficial effects on total sleep time and efficiency, with moderate benefits for sleep quality6. Moderate levels of physical activity can indirectly predict sleep quality over time through emotion regulation60. Furthermore, poor sleep quality exacerbates stress responses of the hypothalamic–pituitary–adrenal (HPA) axis (e.g., cortisol secretion), while physical activity serves as an effective strategy for reducing cortisol levels and improving sleep quality61. This provides a biological explanation for the potential internal mechanisms through which physical activity affects sleep quality. Therefore, active participation in physical activities, including campus sports and leisure exercises, is of great significance for improving sleep quality among college students.

The mediating effect of rumination

This study demonstrated that rumination mediates the relationship between physical activity and sleep quality among college students. This finding supports Hypothesis 2 and aligns with previous research35,36. First, physical activity was found to negatively predict rumination in college students. Rumination exacerbates and prolongs physiological stress responses, including hypothalamic–pituitary–adrenal (HPA) axis activation14. Physical activity facilitates the rapid recovery of the HPA axis following stress induction, particularly in individuals prone to rumination, thereby attenuating heightened cortisol responses62. Regular physical activity may thus serve as a protective factor against the adverse effects of rumination on the HPA axis61. Second, rumination was found to positively predict poor sleep quality among college students. Rumination directs individuals’ attention toward their negative emotions and behaviors, often leading to persistent overthinking about these emotional states and their underlying causes. This cognitive process results in excessive brain activation, ultimately impairing sleep quality63. The sleep cognitive model further suggests that rumination, as a form of intrusive thinking, heightens attention to perceived negative cues before sleep, increasing cognitive wakefulness and making it more difficult to fall asleep64. Notably, rumination remains significantly associated with sleep quality even after controlling for negative mood65. Additionally, rumination may directly impair sleep quality by disrupting immune or endocrine processes66. In summary, these findings highlight rumination as a critical mediator in the relationship between physical activity and sleep quality, offering insights into the psychological, cognitive, and physiological mechanisms underlying this association, and enriching the theoretical understanding of how physical activity influences sleep quality through rumination.

The mediating effect of psychological resilience

This study further revealed that psychological resilience mediates the relationship between physical activity and sleep quality, supporting Hypothesis 3 and aligning with previous research26. On the one hand, physical activity was found to significantly and positively predict psychological resilience. Research indicates that vigorous physical activity is positively correlated with psychological resilience, and resilience scores increase significantly with higher levels of vigorous physical activity67. From a physiological perspective, physical activity reduces cortisol levels (a stress hormone), thereby alleviating stress responses and enhancing an individual’s ability to cope with stress68. Additionally, exercise promotes the release of neurotransmitters such as dopamine, serotonin, and endorphins, which contribute to improved mood, reduced anxiety, and ultimately, strengthened psychological resilience69. Neural mechanisms further reveal that physical activity may enhance psychological resilience by strengthening specific brain regions and large-scale neural circuits70. On the other hand, psychological resilience negatively predicts poor sleep quality among college students. Psychological resilience, characterized as the ability to recover from stress, is essential for maintaining sleep quality71. Individuals with higher psychological resilience are better equipped to relieve stress, mitigate the adverse effects of negative emotions, and improve sleep quality. Moreover, psychological resilience plays a significant role in alleviating neurohormonal changes caused by sleep deprivation, thereby promoting better sleep quality72. In summary, this study validates the mediating role of psychological resilience in the relationship between physical activity and sleep quality. This study provides a comprehensive framework for understanding how physical activity enhances psychological resilience and, in turn, improves sleep quality, offering valuable insights for future research and interventions aimed at promoting well-being among college students.

The chain mediating effect of rumination and psychological resilience

The study contributed to providing evidence for the chain mediating effect of rumination and psychological resilience in the relationship between physical activity and sleep quality among college students. Psychological resilience is influenced by the interaction of internal psychological factors (e.g., cognitive styles and personality traits) and external environmental factors73. As a negative cognitive style, rumination is an internal psychological factor that significantly affects psychological resilience74. On the one hand, physical activity reduces rumination, enhances individuals’ ability to respond to and cope with stress and negative events, and helps alleviate negative emotions in the face of adversity. On the other hand, reducing rumination contributes to improving psychological resilience, thereby promoting individuals’ capacity for psychological recovery75. In summary, this study not only reveals the direct impact of physical activity on sleep quality but also highlights the critical roles of rumination (a cognitive factor) and psychological resilience (a psychological factor) in this relationship. Based on these findings, practical interventions should integrate physical activity, cognitive interventions, and psychological interventions to maximize the improvement of sleep quality among college students.

Limitations and prospects

This study has some limitations that must be acknowledged. Firstly, this study employs a cross-sectional research method, which cannot establish causal relationships between variables. Future research could adopt longitudinal tracking methods to delve deeper into the relationships between physical activity, rumination, psychological resilience, and sleep quality. Secondly, the data in this study were all collected from self-reported questionnaires by college students, a method that may introduce bias and thus affect the accuracy of the research findings. Future studies should consider employing multimodal data collection methods, including wearable device monitoring (e.g., actigraphy), biomarker assays (e.g., cortisol levels), passive digital phenotyping (e.g., smartphone usage patterns), standardized behavioral observations, experimental paradigms, and multi-informant reports to enhance the objectivity and validity of measurements. Thirdly, it is important to note that other factors, such as anxiety, depression, and self-control, may also play significant roles in the relationship between physical activity and sleep quality. These factors were not examined in the current study and warrant further investigation in subsequent research to provide a more comprehensive understanding of the underlying mechanisms. Lastly, while scholars have developed and preliminarily validated the positive and negative rumination scale70, the current study did not differentiate between types of rumination. Future research could examine the distinct roles of positive and negative rumination, as well as their potential joint mediating effects with psychological resilience, in the relationship between physical activity and sleep quality. 这项研究有一些必须承认的局限性。首先，本研究采用的是横断面研究方法，无法建立变量之间的因果关系。未来的研究可以采用纵向跟踪的方法来深入研究身体活动、反刍、心理弹性和睡眠质量之间的关系。其次，本研究的数据均来自大学生自述问卷，这种方法可能会引入偏倚，从而影响研究结果的准确性。未来的研究应考虑采用多模式数据收集方法，包括可穿戴设备监测（例如，活动记录仪）、生物标志物分析（例如，皮质醇水平）、被动数字表型（例如，智能手机使用模式）、标准化行为观察、实验范式和多信息报告，以增强测量的客观性和有效性。第三，重要的是要注意其他因素，如焦虑、抑郁和自我控制，也可能在体育活动和睡眠质量之间的关系中发挥重要作用。这些因素在目前的研究中没有被检查，需要在随后的研究中进一步调查，以提供对潜在机制的更全面的理解。最后，虽然学者们已经开发并初步验证了积极反刍和消极反刍量表72，但目前的研究并没有区分反刍的类型。未来的研究可以检验积极反刍和消极反刍的不同作用，以及它们与心理弹性的潜在联合中介作用，在身体活动和睡眠质量之间的关系。

Conclusion

This study explores the relationships among physical activity, rumination, psychological resilience, and sleep quality. The findings reveal both direct and indirect effects of physical activity on college students’ sleep quality, as well as the mediating roles of rumination and psychological resilience in this relationship. The results suggest that encouraging college students to participate in physical activities (such as diverse sports courses, campus sports events or competitions, leisure-time physical activity, etc.) can not only help reduce rumination but also enhance psychological resilience, thereby providing important practical guidance for improving their sleep quality.