Adolescent students from six high schools in Tongliao City, Inner Mongolia, participated in a questionnaire survey. A total of 826 questionnaires were distributed, and after careful screening, 125 were excluded for clearly insufficient responses, resulting in 701 valid participants, yielding an effective response rate of 84.7%. Details of the valid participants are presented in Table 1. Participants were from several schools in Naihanqi of Tongliao City, with 550 participants (78.5%) from rural areas and 133 participants (19.0%) from urban areas. Among the participants, 187 (26.7%) held class leadership positions, while 493 (70.3%) did not. Regarding the only-child designation, 225 participants (32.1%) were only children, while 464 (66.2%) were not. In terms of family structure, 563 participants (80.3%) came from intact families, 67 (9.6%) from single-parent families, and 56 (8.0%) from blended families. The gender distribution was roughly balanced, with a slight majority of females (51.16%), while males comprised 338 individuals (48.84%). As for grade levels, there were 63 participants (9.0%) in the first year of junior high school, with an average age of about 13 years; 118 participants (16.8%) in the second year, with an average age of 14 years; 144 participants (20.5%) in the third year, with an average age of 15 years; 150 participants (21.4%) in the first year of senior high school, with an average age of approximately 16 years; 143 participants (20.4%) in the second year, with an average age of around 17 years; and 74 participants (10.6%) in the third year, with an average age of about 18 years.

Data collection and entry for this study occurred from November 2023 to January 2024.

Item Packing Note: The following four scales were modeled with latent variables in the analysis of the fitted model in question items packaged three or four adjacent to each other when using AMOS.

This study was approved by the Research Ethics Committee of Inner Mongolia University for Nationalities. After establishing contact with teachers from two middle schools in Naiman Banner, Tongliao City, the researcher introduced the purpose and content of the study. After acquiring their consent, the teachers in the local school used the paper version of the questionnaire to conduct the whole class test in their spare time, and the students completed the questionnaire anonymously. Written informed consent and assent forms were acquired from the participants and their legal guardians/relatives. Participation was voluntary, and confidentiality was guaranteed. The data collectors consisted of trained researchers who ensured the standardization of the data collection process. Completing the self-report questionnaire took approximately 30 min.

Descriptive analysis of the data was conducted using SPSS 25.0 software. The correlations between variables were assessed using Pearson product–moment correlation analysis. Structural equation modeling (SEM) was constructed using AMOS 28.0 software, and the mediation effects along with gender difference tests were examined using the bias-corrected nonparametric percentile Bootstrap method. A 95% confidence interval that does not include zero indicates a significant mediation effect. Statistical significance was defined at p < 0.05.

Using only the questionnaire method in this study may create the problem of common method bias, which was controlled by emphasizing confidentiality and positive and negative scoring during the test administration. A common method bias test was also conducted. Validated factor analysis was used, and the results showed that the one-way model fit was poor: χ/df = 22.46, RMSEA = 0.18, SRMR = 0.13, CFI = 0.64, and TLI = 0.57, which suggests that common method bias is not a severe problem and that the effect of common method bias can be ruled out.

Table 2 presents the means, standard deviations, and correlation coefficients for the various variables. The analysis revealed significant positive correlations among bullying experiences, alexithymia, rumination, and non-suicidal self-injury (NSSI) in pairs. This indicates that the more frequently students experience bullying, the more difficult it is for adolescents to recognize, express, and regulate their emotions. Additionally, the severity of bullying correlates with the severity of adolescents’ NSSI behaviors; those who find it more challenging to identify and express emotions, as well as exhibit more severe alexithymia, demonstrate higher levels of rumination. Furthermore, students who engage in rumination more frequently tend to have a higher occurrence of NSSI. The relationships among these variables support the further testing of the proposed hypotheses Figure 1.

This study used structural equation modeling to test for the chain mediation effect to control measurement error. Variables were analyzed, and it was found that all predictor variables had variance inflation factors below 5, so there was no problem with multicollinearity. The total effect of student bullying on non-suicidal self-injury was significant (β = 0.21, SE = 0.46, t = 5.70, p = 0.000), controlling for gender and school year; secondly, two mediating variables - alexithymia and ruminative thinking - were added to the model to obtain the path model shown in Figure 2. It was found that the fitted model was good: χ/df = 3.11, RMSEA = 0.06, SRMR = 0.04, CFI = 0.97, TLI = 0.96.

Student bullying can all successively affect non-suicidal self-injury through alexithymia and rumination, but there are some differences in their internal mechanisms.

In conclusion, Table 3 shows that the total effect of mediation is significant (b = 0.54, SE = 0.35, 95%CI = [0.14, 1.49]), and the direct effect of bullying on NSSI is significant (β = 0.34, SE = 0.27, 95%CI = [0.03,1.06]). The indirect effect of alexithymia (β = 0.05, SE = 0.04, 95%CI = [0.001,0.18]) and rumination (β = 0.08, SE = 0.05, 95%CI = [0.02,0.23]) was significant. And the indirect effects of alexithymia and rumination as chain mediators were significant (β = 0.07, SE = 0.06, 95%CI = [0.02,0.21]) Figure 3.

The present study examined whether there was cross-sex consistency in the mediating effects of alexithymia and rumination. First, the chained mediation effect models of bullying on non-suicidal self-injury among adolescent students were tested separately for boys and girls. ²²The results showed that the fit indices for the boys’ model were: χ/df = 2.60, RMSEA = 0.07, SRMR = 0.06, CFI = 0.95, TLI = 0.94; and for the girls’ model were: χ/df = 1.92, RMSEA = 0.05, SRMR = 0.05, CFI = 0.97, TLI = 0.96. Overall, all the fit indices were consistent across genders. The fit indices were in the acceptable range for equal group comparisons. Subsequently, the method of multi-group comparison in structural equation modeling was used to set up the equivalent model, Model 1 (baseline model), defining that the boys’ and girls’ groups have the same model structure and the path coefficients are estimated freely; Model 2: restricting the corresponding path coefficients of the boys’ and girls’ groups to be equal based on Model 1. It was found that the fitting results of the two models were acceptable (χ/df = 2.18, RMSEA = 0.04, SRMR = 0.06, CFI = 0.96, TLI = 0.95; χ/df = 4.10, RMSEA = 0.07, SRMR = 0.07, CFI = 0.88, TLI = 0.88); moreover, the difference in fitting indices between the two models ΔCFI and ΔTLI were both greater than 0.01, which indicated significant gender differences (Jiang et al., 2023; Cheung and Rensvold, 2002; Kuo et al., 2024). With both male and female students significantly predicted by their respective chain mediators, the mediation effect value for female students (0.12) was twice as high as that for male students (0.06), suggesting that the chain mediators of rumination thinking and Alexithymia were more predictive for female students than for male students in the relationship between student bullying and non-suicidal self-injurious behaviors.

On this basis, the mediating roles and gender differences of alexithymia and rumination thinking between bullying and non-suicidal self-injury among students were examined. As shown in Table 4, both boys’ and girls’ chain mediation for both alexithymia and rumination thinking were significant as well (Boys:β = 0.06, SE = 0.07, 95%CI = [0.01,0.27]; Female: β = 0.12, SE = 0.21, 95%CI = [0.001,0.71]), and the paths mediated by alexithymia were non-significant; boys’ rumination thinking mediated the relationship between bullying and non-suicidal self-injury as compared to girls’ (β = 0.08, SE = 0.07, 95%CI = [0.01, 0.27]). Separate paths for girls’ alexithymia and rumination were found to be non-significant from the data, and only the overlap of both alexithymia and rumination played a role.

The results of the data indicated that adolescent student bullying significantly and positively predicted their non-suicidal self-injurious behaviors and Hypothesis H1 was confirmed, which is also consistent with previous studies. Several scholars have found a positive correlation between student bullying and non-suicidal behavioral self-injury (Baiden et al., 2017; Jiang et al., 2016; Claes et al., 2015). Some foreign scholars conducted a 1-year observational study on middle school students and found that being bullied by peers significantly predicted individuals’ non-suicidal self-injurious behavior (Jutengren et al., 2011). In addition, Moore et al. concluded that there is sufficient evidence to suggest a causal relationship between bullying and self-injury in childhood and adolescence (Moore et al., 2017). All of these studies provide a favorable basis.

In addition, the findings of this study also fit with the general stress theory and the interpersonal model. On the one hand, the general stress theory suggests that when an individual encounters environmental stimuli or stressful events (stressors), such as being bullied, and is unable to resolve or change their behavior to adapt to the demands of the environment, a subjective feeling of being oppressed is generated in their physiological psychology under such stressful stimuli. Thus, they are likely to alleviate or relieve the feeling of being oppressed through behaviors such as non-suicidal self-injury avoidance (Van Geel et al., 2015). On the other hand, Hilt suggests that non-suicidal self-injurious behaviors are ways of negatively coping with negative interpersonal relationships or events and that individuals experience changes in their interpersonal relationships with peers around them after experiencing a negative event of bullying (Hilt et al., 2008). The last tendency is to escape from difficult situations or to attract others for help through non-suicidal self-injurious behaviors (Muehlenkamp et al., 2013).

This outcome may be due to the rapid physical development of adolescents and their relative psychological immaturity, which makes it difficult for middle school students to adopt effective coping strategies when they are confronted with student bullying as a socially stressful event that cannot be resolved over a long period (Vacca et al., 2023) When the negative emotions are not released for an extended period, students will seek extreme ways to relieve their emotions or seek help, which will gradually strengthen the behavior and form a vicious circle in the long run, thus increasing the frequency of non-suicidal self-injurious behaviors. Therefore, teachers should pay more attention to the emotional and thinking problems of adolescents whom students have bullied, find out in time, and provide effective guidance to prevent bullying from causing more severe consequences.

Findings suggesting that student bullying can indirectly influence adolescents’ non-suicidal self-injurious behaviors through alexithymia are similar to those of Thomassin, who found that emotional expression mediates the relationship between emotional abuse and self-injurious behaviors in childhood (Thomassin et al., 2016). On the one hand, a small number of studies have investigated a positive correlation between alexithymia and bullying victimization (Önal Sönmez et al., 2020). Due to memory fragmentation and attentional lopsidedness in post-victimization individuals (Van der Kolk and Fisler, 1995), they have difficulty effectively utilizing their cognitive abilities. They have difficulty effectively utilizing their cognitive resources to process these emotional states and physiological arousal (Lischke et al., 2022). This may further exacerbate an individual’s difficulty in expressing emotions by causing bullying victims to have difficulty recognizing their own emotions and thus expressing and regulating traumatic emotions (Cowie and Berdondini, 2002). The more severe the student bullying, the higher the level of alexithymia. On the other hand, the traumatic experience serves as an external unpleasant stimulus because of prolonged exposure to the bullying event. Emotions may avoid unpleasant stimuli through non-suicidal self-injurious behaviors, creating negative reinforcement associations (Liu et al., 2022). Meanwhile, empirical studies have demonstrated that alexithymia directly predicts non-suicidal self-injurious behavior (Zhang et al., 2023).

The findings also fit with the social mechanism of alexithymia, which suggests that children’s exposure to certain undesirable social factors during their development, such as socially traumatic events like bullying, may cause severe damage to the individual’s psyche, which in turn reduces their level of socialization. As a result, the emotional regulation ability and self-evaluation system of those who experience bullying may be impaired and exhibit non-adaptive behaviors (Wang and Zhang, 2023).

This study also found that ruminative thinking indirectly influences the relationship between student bullying and adolescent non-suicidal self-injurious behaviors, allowing hypothesis H3 to be confirmed, which is similar to the findings of Qin et al. (2024). Qian Qin concluded that rumination mediates the relationship between major life events and non-suicidal self-injury. On the one hand, the results are consistent with the expectations of the Stress Response Model, which posits that rumination is a subsequent response to a stressful event. When stressful events lead to negative outcomes, individuals tend to keep thinking about these negative outcomes (Robinson and Alloy, 2003). Bullying serves as an ongoing stressor (Kampoli et al., 2017). When individuals are unable to resolve and adapt, it can lead adolescents to become trapped in the negative effects of the bullying incident that are difficult to extricate themselves from, immersing themselves in distressing emotions, i.e., falling into rumination thinking. In the absence of effective adjustment strategies for the individual, compensatory behaviors such as non-suicidal self-injury may be adopted to restore personal and environmental balance in this way (Yusin, 1974; Agnew, 1992). Meanwhile, Malamut confirmed that bullying has a positive predictive effect on rumination thinking (Malamut and Salmivalli, 2021), supporting the present study’s findings.

On the other hand, the findings are consistent with the emotional cascade model, which suggests that ruminating on even small negative stimuli can exacerbate the experience of negative emotions (Selby et al., 2013). Rumination thinking manifests as a sustained focus on the negative stimulus and a repeated experience of the causes and consequences of the negative event. This persistent rumination further exacerbates negative emotions, and strong negative emotions may trigger more rumination, resulting in an emotional cascade (Selby et al., 2008). When rumination reaches a certain level where the individual is unable to cope with the psychological pain, the individual will use non-suicidal self-injury to alleviate their psychological pain (Qin et al., 2024). The following are some examples of non-suicidal self-injury. People report an increase in positive emotions in non-suicidal self-injury, which predicts more severe self-injury tendencies (Hasking et al., 2018), suggesting that non-suicidal self-injury is reinforced, making it a habitual way of coping. Meanwhile, several studies have confirmed that rumination directly predicts non-suicidal self-injury (Selby et al., 2010; Coleman et al., 2022), providing a basis for this study’s findings. Thus, rumination plays a partial mediating role in student bullying and non-suicidal self-injury.

The present study further found that alexithymia and rumination thought chain mediated the effects of student bullying on non-suicidal self-injury, and hypothesis H4 was supported. This finding illustrates that adolescents’ bullying experiences lead to individuals who are prone to difficulties in recognizing and expressing emotions and that individuals with high levels of alexithymia tend to have higher levels of rumination thinking and are ultimately more prone to non-suicidal self-injury. Specifically analyzed, student bullying leads individuals to develop high levels of alexithymia, and according to previous research, alexithymia, and rumination are two typical characteristics of depression (Taylor, 2000). The two are intrinsically congruent, and individuals with alexithymia tend to exhibit higher levels of rumination, which further diminishes their problem-solving and adaptive skills. To alleviate the resulting distress, individuals may engage in non-suicidal self-injurious behaviors.

On the one hand, goal progression theory suggests that when individuals are sensitive to unfulfilled goals, they tend to think about the emotional goals of the negative event repeatedly (Martin et al., 2003). On the one hand, Individuals with alexithymia have relatively little information to guide their behavior in emotional situations due to difficulties in identifying and expressing their own and others’ emotions. As a result, they may tend to think repetitively and analytically about the emotional goals triggered by bullying incidents (Wang and Zhang, 2023), i.e., to fall into rumination thinking, accumulate negative emotions, and ultimately release pain through non-suicidal self-injury. On the other hand, emotion recognition and expression can directly affect interpersonal situations (Kornreich, 2002). Difficulty in identifying and expressing emotions is a distinctive feature of people with alexithymia, which often leads to interpersonal tension, which in turn causes them to reduce their interactions with the outside world passively. In such situations, their thoughts and attention are more focused on internal emotions, and they repeatedly think about less unpleasant targets, i.e., ruminative thinking, which increases the risk of nonsuicidal self-injurious behavior. In addition, adolescents are in a sensitive period of physical and mental development, facing their changes, schooling, and various pressures in life, when they have insufficient ability to deal with problems and frequently experience negative emotions (Stephenson, 1985; Arnett, 1999). Therefore, when faced with bullying, they have difficulty adapting, accepting, or resolving the situation, as well as expressing their emotions. They can only think about the negative results repeatedly, which leads to negative emotions, and then non-suicidal self-injurious behaviors may occur. Moreover, Liu Yuxing and other scholars pointed out that alexithymia positively predicts rumination, and showed that alexithymia and rumination mediated the chain between childhood trauma and suicidal ideation in medical students (Liu et al., 2022). Which is similar to the hypothesis of the present study. Accordingly, affective alexithymia and rumination play a chain-mediating role in the relationship between bullying and self-injurious behaviors among students.

The data results indicate that there are significant gender differences in the chain mediation model regarding the impact of bullying on non-suicidal self-injury (NSSI), thus supporting Hypothesis H5. Specifically, the predictive power of the chain mediation model for females is significantly higher than that for males. Previous studies have shown that females exhibit significantly higher levels of rumination compared to males (Nolen-Hoeksema, 1987; Bugay, 2011). In particular, in response to negative events, females are more likely than males to cope by repeatedly thinking about and focusing on the potential causes of the events. Additionally, because females generally possess higher levels of empathy and emotional understanding (Espelage and Swearer, 2004), they may struggle to escape and effectively regulate these emotions.

However, existing research has indicated that the prevalence of alexithymia is higher among males or shows no significant gender difference (Carpenter and Addis, 2001; Viinikangas et al., 2009). This finding slightly contrasts with the results of the current study. There may be several reasons for this discrepancy:

First, the variations in the selection of variables between previous studies and this research may contribute to the differing findings, as this study included the variable of rumination. Given that different variables can influence gender effects in varying ways and directions, this may lead to conclusions that diverge from those of prior research.

Second, this study examines alexithymia and rumination only in a generalized manner, without delving deeper into the individual dimensions of alexithymia and rumination.

Lastly, this study is limited by its focus on a specific population, as it only surveyed rural adolescents, and the sample size was relatively small, which may weaken its representativeness. Additionally, the research method is confined to a cross-sectional design, lacking in-depth longitudinal analysis, which could explain the discrepancies between these results and those of earlier studies.

In summary, adolescents who experience student bullying are at risk for potential individual physical, mental, and social developmental problems. Bullied individuals may be more inclined to resort to non-suicidal self-injurious behaviors due to chronic psychological stress and unreleased negative emotions. This behavior affects an individual’s physical and mental health and may also negatively affect their social relationships, academic performance, and future development (Ttofi and Farrington, 2011). Student bullying is a significant impediment to an individual’s normal development, as previous research has found that adolescents who are frequently bullied are more likely than those who are not to experience a range of physical, psychological, and behavioral symptoms, including headaches, stomachaches, depression, non-suicidal self-injurious behaviors, and even suicidal behaviors (Rigby, 1998; John Calvin, 2023; Claes et al., 2015; Litwiller and Brausch, 2013). Therefore, it is necessary to deepen the research on the breadth and depth of student bullying to provide theoretical guidance for preventing and intervening in the phenomenon.

Through systematic analysis and evidence-based argumentation, this paper delves into the relationship between bullying experiences and non-suicidal self-injury (NSSI), proposing a new chained mediation model featuring alexithymia and rumination. This model enhances and expands empirical research in general strain theory (GST). According to GST, when individuals encounter stressful stimuli that they are unable to resolve or adapt to, they may experience forced subjective perceptions—such as difficulty recognizing, expressing, or regulating emotions, as well as persistent rumination on negative emotions triggered by stressful events. This intensifies the negative impact of bullying, ultimately leading individuals to maladaptive behaviors as a means to relieve the oppressive experience. By exploring the mediating roles of psychological factors such as alexithymia and rumination, this study enriches the conceptual understanding of the subjective experience of oppression, refines and extends general strain theory, and provides a new cognitive-emotional theoretical perspective and explanatory framework in this domain.

The practical significance of this study is particularly notable within the educational context. It not only aids schools and educators in gaining deeper insights into the roots, manifestations, and impacts of bullying and non-suicidal self-injury (NSSI) behaviors but also provides a scientific basis for developing and implementing effective school intervention or support programs. Through comprehensive research, schools can design more targeted prevention strategies, such as establishing anti-bullying mechanisms, conducting mental health education, and providing psychological counseling and support, thereby fostering a safe, respectful, and inclusive school environment.

First, the study found that bullying successively affects non-suicidal self-injury through alexithymia and rumination, suggesting that socially traumatic experiences can lead to non-adaptive behaviors in bullied students, posing a threat to an individual’s physical and mental health. This effect is both severe and far-reaching, so parents and schools should pay more attention to the physical and mental health of the bullied. Secondly, more mental health education courses should be introduced in schools to popularize anti-bullying knowledge and make students aware of the seriousness of bullying. At the same time, students should be guided to look at themselves correctly, cope with negative events positively, and promote the positive development of emotion regulation and cognitive attribution to enhance adaptive ability. In addition, as different genders and individuals show different ways of coping with bullying among students, we should pay attention to each individual without discrimination and provide personalized counseling according to different situations.

Furthermore, given that bullying is a negative interpersonal event, teachers should always pay attention to interpersonal interactions within the classroom and properly manage interpersonal relationships among students. Instant education should be provided to the bullies, immediate guidance should be given to the bullied, and correct guidance should be given to the bystanders. Adolescent students face multiple pressures and rapid physical and mental development. Schools can organize outdoor activities to help students relax and promote the harmonious development of interpersonal relationships. Finally, previous studies have mostly focused on common emotional problems such as depression, anxiety, and insomnia in bullied students but less on aspects such as alexithymia and rumination. Therefore, by delving into other aspects of cognitive emotions, we can gain a more comprehensive understanding of the process of bullying’s impact on adolescents and provide solid theoretical support for bullying intervention.

Although this study has yielded a series of valuable findings and has made contributions to advancing knowledge in related fields, it also recognizes several significant limitations that need to be addressed in future research endeavors.

Firstly, the study focused exclusively on rural middle school students as the research subjects, which somewhat restricts the generalizability and applicability of the findings. To provide a more comprehensive understanding of the issue, future research could consider including urban middle school students as a comparative control group. This would enrich the content and results of the study, making the conclusions more representative and persuasive.

Secondly, the study employed a cross-sectional design, which, while effective in revealing associations between variables at a single point in time, limits the exploration of the dynamic relationships among variables over time. To gain deeper insights into the intrinsic connections between variables and to confirm the reliability of the findings, future studies should adopt longitudinal tracking methods. This would involve prolonged observation and recording to uncover the changes and interrelations among variables across time.

Thirdly, the evaluation in this study was conducted by teachers, which introduces the possibility of bias, such as social desirability effects. Despite ensuring anonymity and providing training for the teachers to follow standardized procedures, the presence of teachers during the assessment may have influenced students’ responses. This limitation should be considered when interpreting the findings. Future research should aim to minimize this type of bias by employing external assessors or online survey methods to enhance the accuracy and objectivity of the data.

Fourthly, this study did not include an evaluation of clinical diagnoses such as eating disorders, borderline personality disorder, or other psychiatric conditions that are known to significantly increase the risk of non-suicidal self-injury (NSSI). While efforts were made to control for certain psychological variables, the absence of detailed diagnostic data limits the ability to comprehensively interpret the findings. Future research should incorporate structured diagnostic interviews or validated clinical screening tools to better account for the influence of these psychiatric factors and provide a more nuanced understanding of the predictors of NSSI.

Lastly, while this research primarily focused on exploring chain mediation effects, which provides new insights into the relationship between bullying and self-injurious behaviors, the content of the study still requires further expansion. Future investigations could consider including moderation effects to examine how other factors influence the chain mediation process, providing more comprehensive and in-depth theoretical support for interventions targeting bullying. This would in turn help to more effectively prevent and reduce occurrences of adolescent self-injurious behaviors.

Furthermore, future research should not only deepen theoretical exploration but also actively incorporate practical intervention studies as an essential component to further enrich and enhance the research content. By designing and implementing a range of scientifically sound practical intervention measures, it could create a more positive, healthy, and uplifting learning atmosphere for adolescents, effectively promoting their physical and mental well-being. Such practical intervention research would not only validate the effectiveness and feasibility of theoretical hypotheses but also provide valuable practical experience and reference for educators, parents, and policymakers, thus collectively contributing to the healthy growth of adolescents.