Abstract
INTRODUCTION: This study investigates the influence of adult attachment on career decision-making among Chinese university students, focusing on the chain mediation roles of self-differentiation and social support. Grounded in attachment theory and Bowen's family systems theory, the proposed model "adult attachment → self-differentiation → social support → career decision-making" elucidates how emotional dependency indirectly shapes career choices through psychological maturity and social resource integration, offering novel insights for career counseling. METHODS: Data were collected from 908 students across three universities in Shandong Province using standardized scales: the Experiences in Close Relationships Inventory, Revised Differentiation of Self Inventory, Social Support Rating Scale, and Career Decision Inventory. Structural equation modeling (SEM) tested the hypothesized chain mediation, supplemented by cross-group validation and extreme-group analysis to ensure robustness. RESULTS: Structural equation modeling revealed that adult attachment indirectly affected career decision-making through self-differentiation and social support (β = -0.048, p < 0.01), with self-differentiation serving as a partial mediator. Secure attachment enhanced social support acquisition via higher self-differentiation (β = 0.368), ultimately improving decision efficacy (β = 0.330). A direct path (β = -0.059, p < 0.05) confirmed attachment's independent impact. Grade-level differences emerged: mediation effects were pronounced among lower-grade students, while seniors prioritized practical factors. DISCUSSION: The findings validate an emotion-psychology-resource transformation mechanism, demonstrating how secure attachment optimizes decisions through self-differentiation and social support. However, collectivist cultural contexts attenuated self-differentiation's standalone effects. The results advocate integrated interventions combining emotional regulation and resource mobilization, proposing stage-specific guidance: psychological empowerment for early undergraduates and practical resource provision for seniors. Limitations include regional sampling bias and scale cultural adaptability; future cross-cultural longitudinal studies are warranted.