The mediating role of cognitive flexibility in the relationship between social support and non-suicidal self-injury among adolescents: a multicenter cross-sectional study

Chenchen Wang, Jiaqi Zheng, Guanghui Shen, Hong Chen, Xinwu Ye, Chung‐Yi Li, Bin WuFrontiers in Psychology 2025 · DOI

This study, published in Frontiers in Psychology, explores the psychological machinery behind why social support protects adolescents from non-suicidal self-injury (NSSI). By analyzing a large clinical sample, the researchers demonstrate that social support doesn't just provide a safety net—it actually fosters "cognitive flexibility," which in turn helps teens navigate emotional distress without resorting to self-harm.

TL;DR

Social support reduces self-injury in adolescents partly by boosting their cognitive flexibility—the mental ability to switch between thoughts and strategies. This "bridge" effect is significantly stronger in female adolescents, suggesting that social connections may play a unique role in how girls develop adaptive problem-solving skills.

What Problem Does This Paper Address?

Non-suicidal self-injury (NSSI)—deliberate self-harm without suicidal intent—is a global concern, with prevalence rates peaking during the "turbulent" years of mid-adolescence (ages 14–15). While researchers have long known that strong social support (from family and friends) acts as a protective buffer, the specific how has remained fuzzy.

The authors tackle this by integrating the stress-buffering hypothesis with cognitive models. They hypothesize that social support provides a "training ground" for the brain: by exposing teens to diverse perspectives and supportive problem-solving, it enhances their cognitive flexibility. A flexible mind can find alternative ways to cope with pain, whereas a "rigid" mind might see self-injury as the only escape.

Methods

The researchers conducted a multicenter cross-sectional study across 14 psychiatric hospitals and clinics in China.

  • Participants: 2,109 adolescents (ages 12–18).
  • Psychological Scaling: Used the Multidimensional Scale of Perceived Social Support (MSPSS) and the Functional Assessment of Self-Mutilation (FASM).
  • Neuropsychological Testing: Unlike many studies that rely solely on self-reports, this study used the Wisconsin Card Sorting Test (WCST). This is a "gold standard" task where participants must sort cards based on shifting rules (color, shape, or number), providing an objective measure of the brain's ability to adapt to new information.
  • Modeling: They used Structural Equation Modeling (SEM) to map the pathways between support, flexibility, and NSSI, followed by a multi-group analysis to check for gender differences.

Datasets

The study utilized a primary dataset of 2,109 Chinese adolescents recruited from clinical settings between 2021 and 2023. This is a "clinical" dataset, meaning the participants were already engaged with psychiatric services, which differs from general school-based population samples often seen in this field.

Key Insights and Ideas

  • The Mediation Bridge: Cognitive flexibility was confirmed as a partial mediator. It accounted for 8.33% of the total effect social support has on reducing NSSI. While this percentage seems small, it identifies a specific, modifiable cognitive target for therapy.
  • Social Support as Cognitive Training: The paper suggests that social networks do more than offer "hugs"; they offer "perspectives." Interacting with a support system forces an adolescent to process different viewpoints, which exercises the neural circuitry required for cognitive shifting.
  • The Gender Gap in Flexibility: One of the most striking findings was that the link between social support and cognitive flexibility was significant for girls ($\beta = 0.11$) but virtually non-existent for boys ($\beta = 0.02$). The authors suggest this may be due to socialization: girls are often encouraged to process emotions through social dialogue, which may link their social lives more closely to their cognitive development.

Key Results

  • Direct Protection: Higher social support was directly linked to lower NSSI frequency ($\beta = -0.11, p < 0.001$).
  • Cognitive Link: Social support positively predicted cognitive flexibility ($\beta = 0.09, p < 0.001$).
  • NSSI Reduction: Higher cognitive flexibility was associated with lower NSSI ($\beta = -0.08, p < 0.001$).
  • Model Fit: The SEM model showed "great fit" ($\chi^2 = 22.52, CFI = 0.99, RMSEA = 0.047$), indicating the proposed theoretical "bridge" is statistically robust.

Limitations and Future Work

  • The "Snapshot" Problem: As a cross-sectional study, it captures a single point in time. It cannot definitively prove that social support causes cognitive flexibility; it’s possible that more flexible teens are simply better at seeking out support.
  • Clinical Bias: Because the sample came from psychiatric facilities, the results might not perfectly translate to "healthy" adolescents in a general school setting.
  • The "Missing" 91%: Since cognitive flexibility only explained ~8% of the mediation, other factors (like emotion regulation or impulsivity) likely carry the rest of the weight.
  • Future Directions: The authors call for longitudinal studies to track these changes over years and suggest that clinical interventions should combine "social skills training" with "cognitive remediation" to maximize protection against self-harm.

Summary updated

Abstract

This multicenter study provides evidence for the protective roles of social support and cognitive flexibility against NSSI in adolescents. The findings suggest that cognitive flexibility partially mediates the relationship between social support and NSSI, with notable gender differences. These results highlight the importance of enhancing both social support and cognitive flexibility in interventions aimed at reducing NSSI among adolescents.
BibTeX
@inproceedings{wang2025mediating,
  author = {Chenchen Wang and Jiaqi Zheng and Guanghui Shen and Hong Chen and Xinwu Ye and Chung‐Yi Li and Bin Wu},
  title = {{The mediating role of cognitive flexibility in the relationship between social support and non-suicidal self-injury among adolescents: a multicenter cross-sectional study}},
  booktitle = {Frontiers in Psychology},
  year = {2025},
  doi = {10.3389/fpsyg.2025.1546751},
}
Figure