Ecological Systems Theory, proposed by Bronfenbrenner (1986), posits that the family serves as the most influential microsystem impacting individual development. PPC, as a significant family factor, has been closely linked to adolescents’ behavioral development (Scharf & Goldner, 2018). PPC refers to the attempts by parents to control their adolescents’ emotions, thoughts, perspectives, and parent-child relationships. This is achieved through strategies such as inducing guilt, withdrawing love, and exerting authority to regulate and supervise the emotional, cognitive, and behavioral development of their children (Barber, 1996). In the Chinese cultural context, parenting is often embedded in collectivistic values and a strong emphasis on family interdependence and obedience, which may make psychologically controlling practices (e.g., guilt induction, love withdrawal, authority assertion) relatively salient in parent-adolescent interactions. According to Glasser’s Choice Theory, PPC undermines adolescents’ need for autonomy (Glasser, 1999). When adolescents perceive that fulfilling their autonomy needs is more important than maintaining a positive relationship with their parents, they are likely to choose not to comply with parental guidance and may engage in behaviors that help satisfy their needs. Research has shown that in Chinese adolescent populations, particularly among middle and high school students, PPC positively predicts PG (Ji et al., 2024; Wang et al., 2022). Meta-analyses further indicate that PPC consistently predicts adolescents’ PG both cross-sectionally and longitudinally (Nielsen et al., 2020; Lin et al., 2020). However, several important limitations remain in the existing literature. First, most prior studies have focused primarily on the direct association between PPC and PG, paying insufficient attention to the underlying psychological and social mechanisms that may explain this relationship. Second, existing research has typically examined family influences in isolation, with limited consideration of how individual characteristics and peer contexts jointly shape adolescents’ gaming behaviors. Third, much of the available evidence has relied on single-mediator models, which may oversimplify the complex developmental processes through which PPC exerts its influence. Consequently, there is a need for more integrative research that simultaneously considers multiple systems and mechanisms. Grounded in Ecological Systems Theory, the present study seeks to address these gaps by examining how PPC relates to adolescents’ PG through both individual (SC) and peer-level (DPA) pathways. By testing a comprehensive mediation model, this study aims to provide a more nuanced understanding of the processes linking family dynamics to PG among Chinese adolescents.

SC is defined as an individual’ s capacity to independently manage their behavior in alignment with personal or societal standards and to plan for the future (Baumeister et al., 2007). According to SC theory (Hay, 2001), parental upbringing can influence an individual’s SC abilities, which in turn affects their behavioral development. Since psychological control is inherently intrusive, excessive psychological control by parents can undermine adolescents’ sense of autonomy and competence, leading to emotional distres, which negatively impacts the development of SC (Bai et al., 2020; Li et al., 2013). Previous research with adolescents found a significant negative correlation between PPC and SC (Bai et al., 2020; Zhang et al., 2024). Furthermore, an individual’ s level of SC plays a crucial role in shaping their behavior. Adolescents with higher SC are better able to evaluate the potential consequences of their actions and understand the appropriate behaviors in various social contexts, thus exhibiting fewer addictive behaviors (Hu et al., 2025). A longitudinal study by Cudo et al. (2023) indicated that SC can significantly negatively predict PG. Although numerous studies have documented the separate links between PPC and SC, and between SC and PG, relatively few studies have integrated these variables within a single explanatory model. Consequently, the potential mediating role of SC in the relationship between PPC and PG has not been sufficiently examined.

During adolescence, peer influences often overshadow parental influences in shaping behavior (Albert & Steinberg, 2011), making DPA, defined as involvement with peers who deviate from societal norms (Chen et al., 2015), a critical developmental factor. The Choice Theory (Glasser, 1999) provides a framework for understanding the PPC-DPA link. When parents exert psychological control to restrict adolescents’ peer choices, this constraint may inadvertently increase the allure of deviant peers. Empirical findings support this notion, revealing a positive association between PPC and DPA (Tian et al., 2019; Lin et al., 2020, 2022). In turn, Social Group Theory (Harris, 1995) and Peer Pressure Theory (Chadee et al., 2019) explain how affiliation with deviant peers shapes individual behavior. To achieve group integration, adolescents tend to adopt the attitudes and behaviors prevalent in their peer context, potentially leading to externalizing problems such as PG (Chadee et al., 2019). The robust association between DPA and PG has been consistently documented in both cross-sectional and longitudinal research (Lin et al., 2020; Li et al., 2022). Although the direct link between DPA and PG is well established, the familial precursors of DPA, particularly the role of PPC, have received limited empirical attention. This gap underscores the need to examine whether DPA serves as a mechanism linking PPC to adolescent PG.

According to buffering hypothesis, positive individual attributes can attenuate the adverse effects of external risk factors on adjustment (Ellis et al., 2017). In this vein, higher SC reduces youths’ susceptibility to detrimental social contexts: it mitigates the impact of negative family processes (e.g., PPC) and adverse interpersonal experiences such as peer deviance on problem behaviors (Li et al., 2015). Accordingly, adolescents with greater SC are less prone to affiliating with deviant peer groups. However, low SC disposes individuals to seek out antisocial peers and, in turn, to engage in rule-breaking (Gottfredson & Hirschi, 1990). Empirical studies consistently show that adolescents with poorer SC are attracted to groups that endorse hedonistic, effort-averse values and lifestyles, display reduced regard for conventional norms, and are more likely to identify with and join deviance-prone peers (Holt et al., 2012; Liu et al., 2020; Wang et al., 2023). Despite these insights, existing research has rarely examined these processes within an integrated developmental pathway. Most prior studies have investigated either the mediating role of SC or the influence of DPA separately, without considering how deficits in SC may increase adolescents’ vulnerability to deviant peer environments. Consequently, the potential sequential mechanism, from PPC to reduced SC, to increased DPA, and ultimately to PG remains largely unexplored.

Grounded in Ecological Systems Theory and Choice Theory, the present study proposes a chain mediation model (Figure 1) to clarify how PPC relates to adolescents’ PG through two developmentally salient pathways: SC and DPA. By jointly considering the family, individual, and peer contexts, the model aims to delineate the integrated relations among these systems and to illuminate mechanisms underlying the emergence of PG in adolescence. Accordingly, we test three hypotheses:

H1. PPC is positively associated with adolescents’ PG.

H2. SC mediates the association between PPC and adolescents’ PG.

H3. DPA mediates the association between PPC and adolescents’ PG.

H4. SC and DPA act as a chain mediator linking PPC to adolescents’ PG.

A total of 2028 adolescents from three schools in Henan Province, China, participated in this study. Their ages ranged from 13 to 18 years (M = 16.15, SD = 1.646), with 1154 males and 874 females comprising the sample. Participants were recruited using a cluster sampling method and completed paper-based questionnaires during school hours. Both student assent and parental consent were obtained after clarifying the voluntary nature of the study and the confidentiality of responses. All research procedures followed were in accordance with the ethical standards of the World Medical Association (Declaration of Helsinki) for experiments involving human subjects. All conducted procedures received approval from the ethical committee of the Academic Ethics and Moral Committee of Xinxiang Vocational and Technical College (Decision No. 045 of 2025). The research protocol was approved by the authors’ institutional review board. Small incentives were provided upon completion, contributing to a high response rate of 92.2% (2028 valid responses from 2200 distributed questionnaires).

Data were analyzed using SPSS (Version 27.0) in conjunction with the PROCESS macro (Hayes, 2013). Preliminary data screening revealed a minimal proportion of missing values (<2% across all variables). In line with methodological recommendations (Schafer, 1999; Bennett, 2001), listwise deletion was applied, as missing data below 5% is unlikely to introduce substantial bias or reduce statistical power. To assess the potential influence of common method bias, Harman’s single-factor test was initially conducted. Next, descriptive statistics and Pearson correlation analyses were performed to examine the means, standard deviations, and bivariate relationships among all study variables. The hypothesized mediation model was tested using PROCESS Model 6. For inference, 5000 bias-corrected bootstrap samples were used to generate confidence intervals for the indirect effects. Based on prior evidence indicating significant associations between demographic characteristics and PG (André et al., 2020), participants’ gender and age were included as control variables.

To examine the potential influence of common method bias, we applied Harman’s single-factor test (Podsakoff et al., 2003) to all self-reported items. The unrotated factor solution extracted nine factors with eigenvalues > 1. The first factor accounted for only 22.93% of the total variance, falling below the 40% threshold. Thus, common method bias is not a serious concern in the present investigation.

As displayed in Table 1, the correlation analysis revealed several significant associations among the variables. Firstly, a negative correlation emerged between PPC and SC (r = −0.195, p < 0.001). Furthermore, SC was found to be inversely related to both DPA (r = −0.281, p < 0.001) and PG (r = −0.337, p < 0.001). In contrast, a positive relationship was observed between DPA and PG (r = 0.447, p < 0.001).

Table 2 shows the three equations applied to evaluate the chain mediation model. In Eq. (1), PPC predicted SC (β = −0.19, t = −8.66, p < 0.001). Eq. (2) demonstrated that PPC predicted DPA (β = 0.15, t = 6.86, p < 0.001), and SC had a negative effect on DPA (β = −0.26, t = −12.06, p < 0.001). Additionally, gender negatively predicted DPA (β = −0.26, t = −6.20, p < 0.001). In Eq. (3), PPC predicted PG (β = 0.18, t = 9.17, p < 0.001), while SC had a negative effect (β = −0.21, t = −10.28, p < 0.001). DPA was a positive predictor of PG (β = 0.34, t = 16.95, p < 0.001). All 95% confidence intervals did not contain zero.

Table 3 illustrates the direct and indirect effect sizes of the model. The total effect was found to be 0.28, with a 95% confidence interval of [0.2398, 0.3221]. The direct effect of PPC on PG was 0.18, with a 95% confidence interval of [0.1391, 0.2148]. The total indirect effect was 0.10, with a 95% confidence interval of [0.0831, 0.1252]. Specifically, the indirect effect through SC to PG was 0.04 (95% CI: [0.0270, 0.0511]), and the indirect effect through DPA to PG was 0.05 (95% CI: [0.0341, 0.0653]). Additionally, the indirect effect through SC to DPA and then to PG was 0.02 (95% CI: [0.0117, 0.0219]). None of the 95% confidence intervals overlapped zero, confirming the statistical significance of all effects.