Childhood trauma and meaning in life in university freshmen: a chain mediation model and its experimental causal test

Abstract

Background

The formation of meaning in life is a pivotal developmental task during the transition to university, yet its potential undermining by early adverse experiences remains unclear. Building on meaning-making and developmental systems perspectives, this study examined a chain mediating model linking childhood trauma to two dimensions of meaning in life—presence of meaning (MLQ-P) and search (MLQ-S)—through psychological adaptation (PA) and emotion regulation (ER), and further explored its causal architecture using an experimental intervention.

Methods

Cross-sectional data were collected from 2,548 university freshmen across multiple Chinese universities. Participants completed the Childhood Trauma Questionnaire (CTQ), the Meaning in Life Questionnaire (MLQ), the Emotion Regulation Questionnaire (ERQ), and the Psychological Adaptation Subscale (CSPS-B-MH). After data-quality screening, we tested the chain mediating model separately for MLQ-P and MLQ-S using PROCESS Model 6 with 5,000 bias-corrected bootstrap resamples. In Study 2, 74 freshmen were randomly assigned to an adaptation-focused writing intervention or a neutral writing control condition. Pre- and post-test assessments measured psychological adaptation, emotion regulation, and meaning in life. The same analytic strategy was applied to test a chain mediation model.

Results

Study 1 revealed significant negative correlations between childhood trauma and all variables. For MLQ-P, the indirect effect via PA alone dominated (95.7% of total indirect effect), while the chain pathway (CT → PA → ER → MLQ-P) was significant but modest (2.1%). For MLQ-S, a different pattern emerged: the indirect effects via PA alone (56.8%), ER alone (22.1%), and the chain (21.3%) were more balanced. Study 2 showed that the intervention was associated with significant improvements in PA, ER, and total meaning in life. Mediation analysis indicated that the intervention’s association with meaning in life was primarily transmitted through PA alone; the chain pathway was not significant.

Conclusions

Study 1 revealed negative correlations between childhood trauma and all variables. For MLQ-P, childhood trauma was negatively associated with psychological adaptation (PA), PA positively with emotion regulation (ER), and ER positively with MLQ-P. The indirect effect via PA alone dominated (95.7% of total indirect effect), while the chain pathway (CT → PA → ER → MLQ-P) was significant but modest (2.1%). For MLQ-S, the indirect effect via PA alone was smaller (56.8%), while ER alone (22.1%) and the chain pathway (21.3%) played larger roles. Study 2 showed that the adaptation-focused writing intervention was associated with significant improvements in PA, ER, and total meaning in life. Mediation analysis indicated that the intervention's association with meaning in life was primarily transmitted through PA alone; the chain pathway was not significant.

Supplementary Information

The online version contains supplementary material available at 10.1186/s40359-026-04719-8.

Introduction

The formation of a coherent sense of meaning in life is a fundamental component of psychological well-being and a central developmental task during emerging adulthood [65]. For university freshmen, this transitional period is characterized by significant shifts in academic demands, social networks, and personal identity, making the search for and maintenance of meaning both salient and potentially volatile [6, 23, 53]. Within this context, meaning in life—encompassing both the presence of meaning (MLQ-P) and the search for meaning (MLQ-S)—serves as a critical resource, buffering individuals against stress and facilitating successful adaptation to university life [15–17]. These two dimensions are distinct both theoretically and empirically [63]: MLQ-P reflects the degree to which individuals perceive their lives as purposeful and coherent, while MLQ-S reflects an active motivational orientation toward seeking such understanding. For university freshmen, a heightened search for meaning may represent a normative developmental process rather than a psychological deficit [6]. The present study therefore examines these two dimensions separately, avoiding the conflation of a potential risk marker with a potentially adaptive exploratory orientation. Consequently, identifying factors that support or undermine meaning in life during this vulnerable transition is a priority for research in educational psychology and practice in student support services.

Theoretically, two interconnected processes offer a plausible explanatory framework for how childhood trauma (CT) may relate to diminished meaning in life among freshmen. Firstly, psychological adaptation (PA) refers to the capacity to achieve a satisfactory fit between oneself and the demands of a new environment—a broad resource that enables individuals to successfully navigate life transitions [18, 37]. Drawing on cognitive adaptation theory [8], this construct emphasizes how individuals manage external challenges—academic pressures, social integration, and role redefinition—rather than focusing specifically on the modulation of emotional responses. Moreover, psychological adaptation operates on a broader timescale, reflecting an individual’s adjustment trajectory over weeks and months, whereas emotion regulation (ER) arises in response to moment-to-moment affective demands [71]. Childhood trauma, by eroding foundational feelings of safety and predictability, can chronically impair this adaptive capacity [57, 61]. Individuals with poorer psychological adaptation may struggle to integrate into the new university environment, experience lower environmental mastery, and consequently find it more difficult to derive purpose and coherence from their experiences, ultimately resulting in lower levels of meaning in life [55]. Secondly, emotion regulation—the ability to modulate emotional responses—is a more proximal process directly implicated in meaning-making [30, 45]. Trauma is strongly linked to the development of maladaptive emotion regulation strategies (e.g., suppression, rumination) and to deficits in adaptive strategies (e.g., cognitive reappraisal) [31, 39]. Effective emotion regulation allows individuals to process life events, including adverse experiences, and integrate them into a coherent narrative, a process essential for sustaining meaning [55]. Conversely, emotion dysregulation depletes cognitive resources, amplifies distress, and can trap individuals in negative emotional states, thereby impeding their ability to perceive life’s purpose and value [47].

Crucially, these two mechanisms are not mutually independent. A developmental systems perspective suggests they may function sequentially [50]. Impaired psychological adaptation, stemming from early trauma, likely creates a context of chronic stress and poor person–environment fit for freshmen. This context, in turn, places excessive demands on emotion regulation resources and elevates the risk of regulation failure [54]. Thus, childhood trauma may first weaken the broad adaptive resources needed to navigate the university transition (psychological adaptation), which subsequently undermines the specific, day-to-day capacity to manage emotions effectively (emotion regulation), ultimately contributing to a diminished sense of meaning in life. While prior studies have examined these constructs in isolation or in simpler mediation models, research testing an integrated chain mediating pathway—in which broad adaptive capacity sets the stage for specific regulatory processes—remains limited, particularly within the context of the university freshman transition.

Guided by these theoretical and empirical gaps, we propose the following research questions: RQ1, how is childhood trauma associated with meaning in life among university freshmen? RQ2, does psychological adaptation mediate the association between childhood trauma and meaning in life? RQ3, does emotion regulation mediate this association? RQ4, do psychological adaptation and emotion regulation operate in a chain to mediate the association between childhood trauma and meaning in life? We emphasize the correlational nature of our cross-sectional design, which precludes causal inference while allowing for a robust test of the hypothesized theoretical pathways.

Our objectives are, firstly, to establish reliable measurements of all constructs in a large sample of university freshmen; secondly, to test the independent mediating roles of psychological adaptation and emotion regulation; and thirdly, to test the proposed chain mediating model using bias-corrected bootstrapping methods. This study contributes in three primary ways. Firstly, it connects the well-established ACEs literature to the positive psychology outcome of meaning in life, focusing on a developmentally sensitive population (university freshmen) during the critical transition. Secondly, it extends prior work by proposing and testing a theoretically grounded chain mediating model. It examines the indirect pathway through which childhood trauma exerts cascading effects on meaning in life via psychological adaptation and emotion regulation. Finally, by identifying these specific mechanisms, the study provides actionable insights for designing multi-level interventions aimed at fostering meaning in life among trauma-affected students, potentially by strengthening both broad adaptive skills and specific emotion regulation capacities.

Despite the theoretical plausibility of this chain mediating model, prior evidence remains largely cross-sectional and correlational in nature. While existing studies have documented correlations among childhood trauma, psychological adaptation, emotion regulation, and meaning in life, they cannot establish the temporal ordering or causal direction implied by the proposed sequential pathway [50, 55]. Specifically, it remains unclear whether impairments in psychological adaptation lead to difficulties in emotion regulation, which in turn diminish meaning in life, or whether these associations reflect reverse or reciprocal influences. Addressing this gap requires designs that move beyond correlational inference to test the causal architecture of the model.

To this end, the present research adopts a two-study approach. Study 1 employs a large cross-sectional sample of university freshmen (N = 2,548) to estimate the associations among childhood trauma, psychological adaptation, emotion regulation, and meaning in life, and to test whether psychological adaptation and emotion regulation function as sequential mediators in the trauma–meaning in life link. Consistent with best practices for mediation analysis with cross-sectional data, we emphasize that Study 1 provides associational—not causal—evidence, serving to establish the presence and magnitude of the proposed pathways in a representative sample.

Study 2 builds on these findings by employing a randomized controlled experimental design to examine the causal role of psychological adaptation within the chain. If psychological adaptation is indeed a proximal mechanism through which interventions can disrupt the trauma–meaning link, then enhancing adaptation should produce downstream benefits for emotion regulation and, ultimately, meaning in life. We test this hypothesis by randomly assigning freshmen to either an adaptation-focused writing intervention or a neutral writing control condition, and assessing changes in psychological adaptation, emotion regulation, and meaning in life from pre- to post-intervention. This design allows us to examine: (a) whether the intervention successfully manipulates psychological adaptation (manipulation check); (b) whether improvements in adaptation lead to subsequent gains in emotion regulation and meaning in life (cascading effects); and (c) whether the indirect effect of the intervention on meaning in life via psychological adaptation and emotion regulation follows the sequential pattern observed in Study 1 (experimental mediation). By integrating cross-sectional and experimental evidence, this two-study design provides a rigorous test of the chain mediating model while addressing the causal ambiguity inherent in prior work.

Theoretical framework and hypothesis development

Childhood trauma and the erosion of meaning-making foundations

Childhood trauma, encompassing experiences of abuse and neglect, represents a profound disruption to normative development with documented cascading effects on mental health across the life course [28]. From a developmental psychopathology perspective, such early adversity interferes with the formation of secure internal working models, coherent self-schemas, and a sense of the world as predictable and meaningful [13, 68]. The meaning-making model posits that individuals maintain psychological well-being by assimilating life experiences into their global belief systems or by accommodating those systems to align with new experiences [55]. Childhood trauma, as a potent and often inexplicable stressor, frequently violates core assumptions about safety, justice, and self-worth, thereby challenging the individual’s global meaning system [52]. If successfully integrated, such experiences can, in some cases, lead to posttraumatic growth; however, without adequate resources, they more commonly lead to enduring impairments in meaning—a state characterized by a diminished sense of purpose, coherence, and significance in life [1, 65]. For university freshmen, who are actively engaged in reconstructing their identities and life goals, the latent negative self-schemas and disrupted worldviews associated with unresolved childhood trauma may become particularly salient, directly hindering the formation of a positive and stable sense of meaning in life during this critical transition [16, 17, 36]. Consequently, we propose a direct negative association between childhood trauma and both dimensions of meaning in life among this population. As argued in the Introduction, the search for meaning may, during the transition to university, reflect a normative developmental process rather than a deficit [6]. Accordingly, while we anticipate that childhood trauma will be negatively associated with both MLQ-P and MLQ-S, the association with MLQ-S may be weaker. For all hypotheses below, the proposed mediation and chain mediation pathways are expected to operate for both dimensions, but we treat the comparison between MLQ-P and MLQ-S as an exploratory research question.

• H1: Childhood trauma is negatively associated with meaning in life among university freshmen.

Psychological adaptation as a primary mediating mechanism

Psychological adaptation refers to the dynamic process through which individuals adjust their cognitions and behaviors to achieve congruence with new environmental demands, thereby maintaining or restoring psychological equilibrium [18]. It is a core competency for navigating life transitions, such as the transition to university, which requires mastering novel academic challenges, rebuilding social networks, and redefining one’s role [22, 37]. The capacity for effective psychological adaptation is not innate but is scaffolded by early life experiences. Stable, supportive, and predictable caregiving environments foster the development of trust, self-efficacy, and flexible coping strategies—the very foundations of adaptive capacity [4]. Conversely, childhood trauma systematically undermines these foundations by modeling unpredictability, eroding trust in others and oneself, and impairing the development of effective coping repertoires [57]. As a result, individuals with a history of trauma often enter new developmental contexts, such as university, with a diminished reservoir of adaptive resources.

This impairment in psychological adaptation can directly compromise the formation of meaning in life. According to the meaning-making model, the ability to find purpose and coherence is facilitated by a sense of environmental mastery and the successful integration of new experiences [7]. Freshmen with impaired adaptation are more likely to experience the university environment as overwhelming and uncontrollable, hindering their ability to engage in the exploratory behaviors and identity commitments from which meaning is derived [24]. They may become preoccupied with managing daily stressors, leaving limited cognitive and emotional resources for the reflective processes necessary for meaning-making. Therefore, childhood trauma is posited to first weaken psychological adaptation, which in turn reduces the individual’s capacity to generate and sustain meaning in life during the transition to university.

• H2: Psychological adaptation mediates the relationship between childhood trauma and meaning in life among university freshmen.

Emotion regulation as a proximal mediating mechanism

Unlike psychological adaptation, which orients individuals toward external environmental demands, emotion regulation operates on the internal emotional landscape. Emotion regulation—the processes by which individuals influence which emotions they have, when they have them, and how they experience and express them—constitutes a more proximal psychological system intimately involved in daily meaning-making [33]. Effective regulation, particularly the use of adaptive strategies like cognitive reappraisal, allows individuals to manage the affective turbulence of life events, maintain cognitive clarity, and integrate experiences into a coherent personal narrative [21, 45]. Conversely, reliance on maladaptive strategies like expressive suppression or rumination exacerbates distress, consumes cognitive resources, and fragments narrative coherence, thereby obstructing the perception of meaning [10, 29].

The development of emotion regulation competencies is heavily contingent upon early caregiving experiences [26, 78]. Childhood trauma, especially emotional abuse and neglect, disrupts this development by failing to provide consistent modeling, coaching, and validation of emotional experiences [75]. This often results in enduring deficits in emotion regulation, characterized by difficulties in identifying feelings, a propensity for emotional avoidance or over-engagement, and a limited repertoire of regulation strategies [31]. For university freshmen facing a confluence of academic, social, and existential stressors, these pre-existing regulatory deficits can be acutely triggered. The subsequent emotion dysregulation can directly impede meaning in life by trapping individuals in cycles of negative affect, preventing the cognitive elaboration needed to extract purpose from experiences, and undermining the sense of emotional coherence necessary for a unified life narrative [48, 55]. Thus, emotion regulation is hypothesized to serve as a second, more proximal pathway through which childhood trauma impacts meaning in life.

• H3: Emotion regulation mediates the relationship between childhood trauma and meaning in life among university freshmen.

A chain mediating model: from broad adaptation to specific regulation

While psychological adaptation and emotion regulation can function as independent mediators, a developmental systems framework suggests a more nuanced, sequential relationship [76]. We propose that these constructs form a chain mediating pathway. Psychological adaptation represents a broader, higher-order capacity to achieve person-environment fit. When this capacity is compromised by childhood trauma, the individual is placed in a state of chronic or recurrent stress due to poor adjustment to the new university environment. This state of poor adaptation, characterized by feelings of alienation, academic overwhelm, and social disconnection, creates a context that persistently taxes the individual’s emotion regulation system [54]. In other words, struggling to adapt to the broader university environment depletes the psychological resources required for effective emotional micro-management [27]. This temporal and functional hierarchy—in which the broader condition of person–environment fit sets the context for the more specific process of emotion regulation—is the core theoretical rationale for specifying PA as the first-stage mediator and ER as the second-stage mediator. Unlike reciprocal or reverse models, this cascade posits that one must first navigate the environment before one can effectively regulate the emotions evoked within it [20]. Together, meaning-making theory [55] and developmental systems theory [50] jointly generate this chain model: the former specifies why early adversity undermines the global meaning system, while the latter specifies how this disruption cascades through successive levels of psychological functioning—from broad adaptive capacity to specific regulatory processes—before culminating in diminished meaning. Consequently, impaired psychological adaptation leads to more frequent and severe episodes of emotion dysregulation.

This sequential impairment culminates in reduced meaning in life. The inability to regulate emotions effectively, itself a consequence of broader adaptive failure, then directly interferes with the meaning-making process. Therefore, the full impact of childhood trauma on meaning in life is not merely the sum of two separate paths but may also flow through a cascading chain: from early trauma to impaired psychological adaptation, which exacerbates difficulties in emotion regulation, which ultimately diminishes the sense of meaning in life. This chain model specifies the functional cascade and theoretical precedence of the mediators.

• H4: Psychological adaptation and emotion regulation play a chain mediating role between childhood trauma and both dimensions of meaning in life among university freshmen. Specifically, childhood trauma predicts reduced psychological adaptation, which in turn predicts impaired emotion regulation and ultimately lower levels of meaning in life. Given the differential nature of MLQ-P and MLQ-S, we further explore whether the chain mechanism operates comparably across the two dimensions.

The hypotheses outlined above (H1–H4) will be tested in Study 1 using a large cross-sectional sample of university freshmen. These hypotheses specify the expected pattern of associations among childhood trauma, psychological adaptation, emotion regulation, and meaning in life, but do not imply causal ordering due to the cross-sectional design.

To address the question of causality and examine whether psychological adaptation indeed functions as a proximal mechanism in the chain, Study 2 employs a randomized controlled experimental design. Based on the same theoretical framework, we propose the following additional hypotheses:

• H5: Compared to the control condition, participants in the adaptation-focused writing intervention will show greater improvements in psychological adaptation from pre- to post-intervention (manipulation check).

• H6: The intervention will lead to greater improvements in emotion regulation and meaning in life compared to the control condition.

• H7: The effect of the intervention on meaning in life will be sequentially mediated by improvements in psychological adaptation and emotion regulation, consistent with the chain mediating pattern hypothesized in H4 (i.e., intervention → psychological adaptation → emotion regulation → meaning in life).

Conceptual model

Figure 1 depicts the core theoretical model underlying the present research. The model specifies a chain mediating pathway through which a distal antecedent (childhood trauma) is proposed to influence meaning in life via two sequential mechanisms: psychological adaptation and emotion regulation. Specifically, the model includes: (a) a direct path from childhood trauma (CT) to meaning in life (MIL); (b) an indirect path through psychological adaptation (PA) alone (CT → PA → MIL); (c) an indirect path through emotion regulation (ER) alone (CT → ER → MIL); and (d) a chain mediating path through both variables in sequence (CT → PA → ER → MIL). This integrated model aligns with developmental systems theory by outlining how a distal risk factor (childhood trauma) influences a positive developmental outcome (meaning in life) through a hierarchy of interconnected psychological processes, spanning from broad adaptive functioning to specific regulatory skills [50, 55]. In the present study, MIL is empirically examined as two separate dimensions: the presence of meaning (MLQ-P) and the search for meaning (MLQ-S), consistent with the theoretical distinction outlined in the Introduction.

Fig. 1.

Conceptual model of the chain mediating relationship. Note. In Study 1, the model is tested with CT as the predictor, and MIL is examined separately as MLQ-P and MLQ-S. In Study 2, the model is tested by experimentally manipulating PA, with the expectation that intervention-induced changes in PA will propagate through ER to MIL

This conceptual model guides both studies reported in this paper, albeit with different operationalizations of the initial predictor. In Study 1, we test the model as depicted, examining whether childhood trauma (measured via retrospective self-report) relates to meaning in life through the proposed sequential mediators. In Study 2, we experimentally manipulate the first mediator—psychological adaptation—through a randomized controlled writing intervention. If the chain model is causally valid, enhancing psychological adaptation should produce downstream improvements in emotion regulation and, ultimately, meaning in life, thereby replicating the sequential pattern observed in Study 1. Thus, while Study 1 tests the presence of the proposed pathways in a naturalistic setting, Study 2 tests their causal validity by intervening at the initial stage of the chain.

Study 1: cross-sectional survey

Aim and overview

Study 1 aimed to test the hypothesized chain mediating model in a large sample of university freshmen using a cross-sectional design. Specifically, we examined whether childhood trauma relates to meaning in life through the sequential mediators of psychological adaptation and emotion regulation, as proposed in Hypotheses 1–4. Consistent with best practices for mediation analysis with cross-sectional data, we interpret the findings in correlational terms—they establish the presence and magnitude of the proposed pathways but do not imply causal ordering. The results of Study 1 provide the empirical foundation for the experimental causal test in Study 2.

Method

Design and context

We used a cross-sectional questionnaire design in universities across China to estimate associations (not causal effects) among childhood trauma, psychological adaptation, emotion regulation, and meaning in life. All scales were administered using validated Chinese versions, and a brief pilot study was conducted to ensure the clarity of scale items. To limit common-method bias, we varied the response anchors of the scales, randomized item order, and embedded attention checks; post-hoc, we inspected single-factor dominance. Data were collected via coordinated online and proctored offline sessions.

Ethics approval and informed consent were obtained. All participants provided written informed consent before completing the survey, which outlined the voluntary nature of participation, confidentiality, and the potential sensitivity of childhood trauma items.

Participants and sampling

The target population was full-time undergraduates from comprehensive and specialized universities. Stratified sampling by university type and major category (e.g., engineering/tech, social sciences, management/economics, health/education, arts/humanities) guided proportional recruitment through class rosters and official channels, with moderate oversampling of underrepresented strata. A total of 3,478 questionnaires were distributed. After excluding invalid responses based on consistency checks, lie-detector items, anomalous completion times, and data completeness criteria, 2,548 valid cases were retained (73.3% valid response rate). The sample consisted of 1,172 men (45.9%) and 1,379 women (54.1%); age M = 18.65 (SD = 1.13); Urban–rural background and SES proxies (parental education, perceived family economic standing) were recorded for descriptive and robustness analyses; primary models were adjusted for gender and age.

Procedures and ethical considerations

Recruitment was coordinated with student affairs offices and program leaders at participating universities in China. Invitations were distributed via official learning portals and course groups (online arm) and through brief in-class announcements followed by proctored survey sessions (offline arm). Before accessing the questionnaire, all students viewed a concise consent page describing aims, voluntary participation, approximate duration, potential sensitivities (trauma items), confidentiality, and the right to skip any question or withdraw at any time without penalty. Proceeding to complete the survey constituted informed consent; no identifying information (names, student IDs, IP addresses) was collected, and responses were de-identified at capture.

Because the survey included childhood-trauma items, additional safeguards were implemented. Participation was strictly voluntary with item-level skipping enabled throughout. The opening page and a persistent footer listed campus counseling services and regional helplines; proctored sessions ended with a brief reminder on how to access support. Traumatic-content items were grouped with neutral buffers and accompanied by a content notice (“some items concern difficult childhood experiences”). For offline administration, proctors were trained to respond to distress and provide referral cards. Data were stored on an encrypted server with access restricted to the research team; data exports used random IDs only.

The study protocol received approval from the university Institutional Review Board/Research Ethics Committee (approval number XHU-260403). Procedures complied with the Declaration of Helsinki and local regulations for human subjects research.

Measures

Childhood Trauma Questionnaire (CTQ)

The CTQ was originally developed by Bernstein [11] and has been widely used in China following revisions by Chinese scholars [77]. The 28-item scale (including 3 validity items) comprises five dimensions: Emotional Neglect (EN), Emotional Abuse (EA), Physical Abuse (PA), Sexual Abuse (SA), and Physical Neglect (PN). Items are rated on a 5-point Likert scale, with higher scores indicating more severe childhood trauma. In this study, the scale demonstrated an internal consistency Cronbach’s α coefficient of 0.852. Subscale reliabilities were as follows: Emotional Abuse = 0.716, PA = 0.833, SA = 0.869, EN = 0.767, PN = 0.742, comparable to those reported in prior Chinese college student samples [77].

Meaning in Life Questionnaire (MLQ)

The MLQ, developed by Steger et al. [64] and revised for Chinese populations [46], consists of 9 items divided into two subscales: Presence of Meaning (MLQ-P) and Search for Meaning (MLQ-S). It uses a 7-point Likert scale. Given that the two subscales are theoretically and empirically orthogonal [63], they were analyzed separately in the present study. In this study, Cronbach’s α was 0.847 for MLQ-P, 0.784 for MLQ-S, and 0.796 for the total scale. These reliability values are consistent with prior Chinese college student samples [46].

Emotion Regulation Questionnaire (ERQ-CRV)

The Chinese Revised Version of the ERQ [72], originally by Gross and John [29], was used. It includes two subscales: Cognitive Reappraisal (6 items) and Expressive Suppression (4 items). A 7-point Likert scale is used, with higher scores indicating more frequent use of the regulation strategy. In this study, Cronbach’s α coefficients for the Cognitive Reappraisal and Expressive Suppression subscales were 0.817 and 0.715, respectively. Cronbach’s α for the total scale was 0.700 in this study. Thus, the ERQ captures the micro-level, intra-individual strategies individuals use to modulate their internal affective states in specific situations [30].

Concise College Student Psychological Quality Scale (Health Version) – Adaptation Subscale (CSPS-B-MH)

This subscale, originally developed by Chinese scholars to assess psychological qualities among Chinese college students, was adopted. It serves as a multidimensional adaptability measure [79] and has been widely used among Chinese college students [74]. We chose this scale over general psychological well-being measures because it was specifically developed to assess psychological adaptation—defined as person–environment fit during life transitions—rather than broad well-being or mental health, closely aligning with our conceptualization of PA [18]. Additionally, as a tool developed and validated within the Chinese higher education context, it captures culturally unique student adaptation concerns absent in translated Western instruments (e.g., integration into the collectivistic campus environment, adjustment to family expectations) [73]. The scale comprises 18 items, each rated on a 5-point Likert scale. Higher scores indicate stronger psychological adaptability. In this study, Cronbach’s α was 0.873. Subscale α were: Vocational Adaptation = 0.841, Social = 0.822, Physiological = 0.746, Academic = 0.802, Interpersonal = 0.729, Emotional = 0.759. These values are consistent with those reported in the scale’s original validation study [73]. In contrast to the ERQ, this scale was selected to assess the macro-level, environment-oriented capacity for achieving person–environment fit during the university transition [20].

Data preparation

Data-quality screening followed prespecified rules. Cases were excluded for: (a) failing any instructed-response/attention check; (b) excessive missingness (> 20% across the battery or > 40% on any single scale); (c) long-string (“straight-lining”) of identical responses ≥ 8 on a multi-item block; (d) implausible completion time (below the 1 st percentile of the offline completion time distribution or < ~ 5 min online); and (e) CTQ minimization/denial flags combined with contradictory response patterns. Remaining item-level missingness was minimal; primary analyses used listwise deletion for PROCESS models to match bootstrap assumptions, and descriptive statistics used available-case analysis.

Power analysis

With N = 2548, the study was well powered for small main effects and typical interaction sizes in psychosocial research. PROCESS bootstrap procedures provided adequate sensitivity for indirect effects of the magnitude observed in the present study; Monte-Carlo checks indicated ≥ 0.80 power to detect small-to-moderate indirect paths and small interactions under our variance structure.

Analytic strategy

All statistical analyses were conducted using IBM SPSS 25.0, with the PROCESS macro (version 3.4; [32]) employed to test mediation effects (Model 4) and moderated mediation effects (Model 6). The Bootstrap method with 5,000 resamples was applied to test the significance of mediation effects, with 95% confidence intervals (CIs). Statistical significance was set at p < 0.05.

Results

Sample characteristics

The analytic sample comprised 2548 undergraduates from universities in China (valid response rate: 73.30%). Participants were 45.90% men (n = 1172) and 54.10% women (n = 1379), mean age 18.65 years (SD = 1.13). Regarding family location, 1439 students (56.40%) were from rural areas and 1112 students (43.60%) were from urban areas. Parental education (highest) was distributed as follows: 682 students (26.77%) had parents with educational attainment below middle school, 1100 students (43.17%) had parents with high school as the highest educational level, and 766 students (30.06%) had parents with college or above. Urban–rural background and SES proxies (parental education, perceived family economic standing) were recorded for descriptive and robustness analyses. Full demographics appear in Table 1.

Table 1.

Demographic and background characteristics (N = 2548)

Characteristic	Category	n	Rate
Gender	Men	1172	45.90%
	Women	1379	54.10%
Age	Mean (SD)	18.62 (1.13)	—
Urban–rural background	Urban	1112	43.60%
	Rural	1439	56.40%
Parental education (highest)	≤ Middle school	682	26.77%
	High school	1100	43.17%
	College or above	766	30.06%

Common method bias checks

Harman’s single-factor test was conducted via exploratory factor analysis. Results showed 17 factors with eigenvalues greater than 1, with the first factor accounting for 17.86% of the total variance, below the 40% threshold, indicating no severe common method bias [58].

Descriptive statistics and correlation analysis

Descriptive statistics and correlations are presented in Table 2. Childhood trauma was significantly negatively correlated with both MLQ-P (r = −0.30, p < 0.01) and MLQ-S (r = −0.15, p < 0.01), as well as with psychological adaptation (r = −0.37, p < 0.01) and emotion regulation (r = −0.11, p < 0.01). MLQ-P was positively correlated with psychological adaptation (r = 0.61, p < 0.01) and emotion regulation (r = 0.19, p < 0.01). MLQ-S was positively correlated with psychological adaptation (r = 0.18, p < 0.01) and emotion regulation (r = 0.29, p < 0.01). Psychological adaptation and emotion regulation were positively correlated (r = 0.16, p < 0.01).

Table 2.

Descriptive statistics and correlations

Variable	1	2	3	4	5	6	7
Covariates
1.Gender	—
2.Age	−0.17**	—
Core Variables
3.Childhood Trauma	−0.01	0.12**	—
4. MLQ-P	−0.04	0.04	−0.30**	—
5. MLQ-S	0.04*	−0.14**	−0.15**	0.22**	—
6.Psychological Adaptation	0.02	0.01	−0.37**	0.61**	0.18**	—
7.Emotion Regulation	−0.11**	−0.02	−0.11**	0.19**	0.29**	0.16**	—
M	—	18.65	46.40	24.05	20.24	61.24	44.51
SD	—	1.13	8.22	5.69	4.49	9.92	6.90

MLQ-P = presence of meaning; MLQ-S = search for meaning. The same below.

^*p < 0.05

^**p < 0.01

^***p < 0.001

Mediating roles of psychological adaptation and emotion regulation (H2 and H3)

Before testing the chain mediation model, we first examined the independent mediating roles of psychological adaptation and emotion regulation using PROCESS Model 4, with separate analyses for MLQ-P and MLQ-S. All models controlled for gender and age. Results are summarized in Table 3.

Table 3.

Simple mediation analyses for MLQ-P and MLQ-S

Path	MLQ-P			MLQ-S
	β	SE	95% CI	β	SE	95%CI
Panel A: PA as Mediator
CT → PA	- 0.38***	0.02	[- 0.50,—0.41]	- 0.38***	0.02	[- 0.50,—0.41]
PA → MIL	0.58***	0.01	[0.52, 0.62]	0.14***	0.01	[0.09, 0.17]
Direct effect	−0.09***	0.01	[- 0.50,—0.41]	−0.09***	0.01	[- 0.13,—0.04]
Total effect	−0.21***	0.01	[- 0.50,—0.41]	−0.08***	0.01	[- 0.10,—0.05]
Indirect effect	−0.15***	0.01	[- 0.50,—0.41]	−0.03***	0.01	[- 0.04,—0.02]
Ratio of indirect	71.3%			39.0%
Panel B: ER as Mediator
CT → ER	- 0.11***	0.02	[- 0.13,—0.06]	- 0.11***	0.02	[- 0.13,—0.06]
ER → MIL	0.16***	0.02	[0.10, 0.21]	- 0.29***	0.01	[0.21, 0.31]
Direct effect	- 0.29***	0.02	[- 0.33,—0.22]	- 0.11***	0.01	[- 0.14,—0.07]
Total effect	- 0.38***	0.02	[- 0.24,—0.18]	- 0.08***	0.01	[- 0.10,—0.05]
Indirect effect	- 0.01***	0.003	[- 0.02,—0.01]	- 0.02***	0.004	[- 0.03,—0.01]
Ratio of indirect	5.7%			22.7%

CT Childhood trauma, PA Psychological adaptation, ER Emotion regulation

The same below. Ratio = indirect effect/total effect

As shown in Table 3, Panel A, psychological adaptation significantly mediated the relationship between childhood trauma and both meaning dimensions. For MLQ-P, the indirect effect was—0.15, 95% CI [- 0.17,—0.13], accounting for 71.3% of the total effect. For MLQ-S, the indirect effect was notably smaller at—0.03, 95% CI [- 0.04,—0.02], accounting for 39.0% of the total effect. Panel B shows that emotion regulation also functioned as a significant mediator for both dimensions, but with the opposite pattern: the indirect effect was modest for MLQ-P (- 0.01, 95% CI [- 0.02,—0.01], 5.7% of total effect) and comparatively larger for MLQ-S (- 0.02, 95% CI [- 0.03,—0.01], 22.7% of total effect). These divergent patterns indicate that psychological adaptation is the dominant mediator for the presence of meaning, whereas emotion regulation plays a relatively more important role for the search for meaning. Both H2 and H3 were supported for both dimensions. We therefore proceeded to the chain mediation model to examine whether these two mediators function sequentially.

Chain mediation analysis with MLQ-P as outcome (H4)

After controlling for gender and age, we examined the chain mediation model with MLQ-P as the dependent variable. The path coefficients are presented in Table 4. Childhood trauma negatively predicted psychological adaptation (β = − 0.38, p < 0.001), which positively predicted emotion regulation (β = 0.14, p < 0.001), which in turn positively predicted MLQ-P (β = 0.09, p < 0.001). The direct effect of childhood trauma on MLQ-P remained significant (β = − 0.08, p < 0.001).

Table 4.

Path coefficients for the chain mediation model with MLQ-P as Outcome

Predictor	PA		ER		MLQ-P
	β	SE	β	SE	β	SE
CT	- 0.38***	0.02	- 0.06***	0.02	- 0.08***	0.02
PA			0.14***	0.02	0.57***	0.02
ER					0.09***	0.02
R2	0.14		0.04		0.39

Table 5 presents the completely standardized indirect effects. The total indirect effect was −0.22, 95% CI [−0.25, −0.20], accounting for 72.8% of the total effect. Psychological adaptation alone (Ind1: CT → PA → MLQ-P) was the dominant pathway, with an effect of −0.21, 95% CI [−0.24, −0.19], representing 95.7% of the total indirect effect. Both the emotion regulation pathway (Ind2: CT → ER → MLQ-P; effect = − 0.01, 95% CI [−0.01, −0.001]) and the chain pathway (Ind3: CT → PA → ER → MLQ-P; effect = − 0.01, 95% CI [−0.01, −0.001]) were significant but modest, contributing 2.2% and 2.1% of the total indirect effect, respectively. The full model explained 39.0% of the variance in MLQ-P.

Table 5.

Indirect effects for the chain mediation model with MLQ-P as outcome

Indirect Path	Effect	SE	95% CI	Ratio
Total indirect	−0.22	0.01	[−0.25, −0.20]	72.8%
Ind1: CT → PA → MLQ-P	−0.21	0.02	[−0.24, −0.19]	95.7%
Ind2: CT → ER → MLQ-P	−0.01	0.001	[−0.01, −0.001]	2.2%
Ind3: CT → PA → ER → MLQ-P	−0.01	0.001	[−0.01, −0.001]	2.1%

Chain mediation analysis with MLQ-S as outcome (H4)

The same chain mediation model was tested with MLQ-S as the dependent variable. Standardized path coefficients are shown in Table 6 As before, childhood trauma negatively predicted psychological adaptation (β = −0.38, p < 0.001), which positively predicted emotion regulation (β = 0.14, p < 0.001). However, the final stage of the model differed markedly: emotion regulation was a stronger predictor of MLQ-S (β = 0.27, p < 0.001) than was psychological adaptation (β = 0.10, p < 0.001), in contrast to the MLQ-P model where PA (β = 0.57) far outweighed ER (β = 0.09).

Table 6.

Path coefficients for the chain mediation model with MLQ-S as outcome

Predictor	PA		ER		MLQ-S
	β	SE	β	SE		β	SE
CT	- 0.38***	0.02	- 0.06***	0.02		- 0.07***	0.02
PA			0.14***	0.02		0.10***	0.02
ER						0.27***	0.02
R2	0.14		0.04			0.13

Table 7 presents the completely standardized indirect effects. The total indirect effect was −0.07, 95% CI [−0.09, −0.05], accounting for 49.9% of the total effect. Notably, the three indirect pathways were more evenly distributed than in the MLQ-P model: Ind1 (CT → PA → MLQ-S) contributed 56.8%, Ind2 (CT → ER → MLQ-S) contributed 22.1%, and Ind3 (CT → PA → ER → MLQ-S) contributed 21.3%. All three were significant. The full model explained 12.7% of the variance in MLQ-S.

Table 7.

Indirect effects for the chain mediation model with MLQ-S as outcome

Indirect Path	Effect	SE	95% CI	Ratio
Total indirect	−0.07	0.01	[−0.09, −0.05]	49.9%
Ind1: CT → PA → MLQ-S	−0.04	0.01	[−0.06, −0.02]	56.8%
Ind2: CT → ER → MLQ-S	−0.02	0.01	[−0.03, −0.001]	22.1%
Ind3: CT → PA → ER → MLQ-S	−0.01	0.003	[−0.02, −0.01]	21.3%

Supplementary analyses

We also examined whether the pattern of mediation held when including additional covariates (e.g., urban–rural background, parental education). The results remained substantively unchanged, indicating the robustness of the findings.

Brief discussion

Study 1 tested the hypothesized chain mediating model in a large sample of university freshmen. The results supported all four hypotheses. Notably, decomposing meaning in life into MLQ-P and MLQ-S revealed distinct patterns: the chain mediation model was substantially more powerful in explaining the presence of meaning (R² = 0.39) than the search for meaning (R² = 0.13), and psychological adaptation dominated the indirect pathways for MLQ-P, whereas emotion regulation played a more prominent role for MLQ-S.

These cross-sectional findings provide associational evidence for the proposed pathways in a representative sample, although they cannot establish their temporal ordering or causal direction. However, the correlational design precludes conclusions about temporal order or causality. It remains unclear whether psychological adaptation truly plays a causal role in shaping emotion regulation and meaning in life, or whether the observed associations reflect unmeasured confounds or reverse causality. To address this gap, Study 2 employed a randomized controlled intervention to experimentally manipulate psychological adaptation and examine its downstream effects on emotion regulation and meaning in life, thereby testing the causal architecture of the chain model.

Study 2: randomized controlled intervention

Aim and overview

Study 1 provided cross-sectional evidence for the hypothesized chain mediating model, demonstrating that childhood trauma relates to lower meaning in life through the sequential mediators of psychological adaptation and emotion regulation. Psychological adaptation emerged as the dominant pathway, accounting for nearly 60% of the total indirect effect. However, the correlational design precludes causal inference regarding the directionality of these associations.

To address this gap, Study 2 employed a randomized controlled experiment to test the causal role of psychological adaptation within the chain model. Drawing on established paradigms of expressive writing [56], best possible selves writing [41], and coping flexibility [19], we designed a brief writing task aimed at strengthening freshmen’s psychological adaptation to university life. This design allowed us to test whether experimentally enhancing psychological adaptation produces downstream improvements in emotion regulation and meaning in life (H5–H7), thereby providing causal evidence for the sequential pathway identified in Study 1.

Method

Design and context

We employed a 2 (group: experimental vs. control) × 2 (time: pre-test vs. post-test) mixed experimental design, with group as a between-subjects factor and time as a within-subjects factor. The experiment was conducted in classrooms at a comprehensive university in Sichuan, China. Participants first completed the pre-test measures on paper. They were then asked to scan a QR code using their mobile phones, which directed them to the Wen Juan Xing platform. The platform automatically and randomly assigned each participant to either the experimental or control condition. Participants completed the writing task on paper according to their assigned condition. Immediately after finishing the writing task, they completed the post-test measures, again on paper. The study was presented as an investigation of “University Student Growth Questionnaire,” and participants were unaware of the experimental hypotheses. The procedure was approved by the institutional ethics committee (approval number XHU-260403), and all participants provided written informed consent prior to participation.

Participants and sampling

An a priori power analysis was conducted using G-Power 3.1 [25] for a 2 (group: experimental vs. control) × 2 (time: pre-test vs. post-test) repeated-measures ANOVA. Assuming a medium effect size (f = 0.25, equivalent to ηₚ² = 0.06), α = 0.05, power = 0.80, and a correlation of 0.5 between repeated measures, the analysis indicated a required total sample size of 34. To account for potential attrition (estimated at 10%) and to ensure adequate power for subsequent mediation analyses (which typically require larger samples), we aimed to recruit 70 participants (35 per group).

Freshmen were recruited through campus advertisements and announcements in introductory psychology courses. A total of 85 students volunteered to participate. Of these, 11 were excluded because they either failed to complete the writing task (writing fewer than 200 Chinese characters), provided responses that were clearly off-topic (e.g., describing unrelated content), or had excessive missing data in the questionnaires. The final sample consisted of 74 first-year students, with 37 randomly assigned to the experimental group and 37 to the control group. The sample included 11 men (14.9%) and 63 women (85.1%), with a mean age of 18.47 years (SD = 0.62). All participants were native Mandarin speakers and had normal or corrected-to-normal vision. Based on self-report at the time of recruitment, no participant reported a history of psychiatric or neurological disorders.

Intervention procedure

The experiment was conducted in group settings and followed a three-step procedure. First, participants completed a paper-based pre-test assessing demographic variables, psychological adaptation, emotion regulation, and meaning in life (approximately 10 min). Second, they scanned a QR code with their mobile phones, which directed them to a Wen Juan Xing page that randomly assigned them to either the experimental or control condition. Third, based on their assignment, participants received a second paper booklet containing the writing instructions and post-test measures.

The experimental group received the following instruction (see Appendix A for full materials):

• “Entering university represents a significant life transition that requires adaptation to a new environment. Reflecting on how one has navigated challenges and identifying one’s adaptive resources can be helpful. Please reflect on the adaptation-related challenges you have encountered since starting university and write at least 400 characters, structured as follows: (1) describe specific challenges and how you initially coped; (2) describe any successful adaptation experiences, however small; and (3) based on what you have learned, outline strategies for handling similar challenges in the future.”

The control group received the following instruction (see Appendix B for full materials):

• “Please describe your daily activities over the past two days in chronological order, in the form of a factual daily log. Include details such as your wake-up time, what you ate, which classes you attended, and your activities during breaks and in the evening. Do not include any subjective evaluations, emotional descriptions, or personal reflections—focus solely on objective facts. Write at least 400 characters, divided into Day 1 and Day 2.”

Both groups wrote for at least 20 min and then immediately completed the paper-based post-test, which was identical to the pre-test.

Measures

All three constructs—psychological adaptation, emotion regulation, and meaning in life—were measured using the same instruments as in Study 1. The pre‑test and post‑test used identical scales, and the order of items within each scale was randomized to reduce memory effects. Demographic variables (gender, age) were collected at the beginning of the pre‑test. The internal consistency of the three measures was satisfactory: Cronbach’s α for psychological adaptation was 0.620 at pre-test and 0.88 at post-test; for emotion regulation, α was 0.674 at pre-test and 0.657 at post-test; for meaning in life, α was 0.672 at pre-test and 0.699 at post-test.

Data preparation

All data were entered and verified for accuracy prior to analysis. The final sample (N = 74) had complete data on all measures; therefore, no missing data imputation was required. Univariate outliers were examined via boxplots for all continuous variables (PA-pre = Pre-test psychological adaptation, ER-pre = Pre-test emotion regulation, MIL-pre = Pre-test meaning in life, PA-post = Post-test psychological adaptation, ER-post = Post-test emotion regulation, MIL-post = Post-test meaning in life), and no extreme values were detected. Normality assumptions were assessed through skewness and kurtosis values. For all variables, skewness ranged from −0.78 to −0.02, and kurtosis ranged from −0.77 to 1.23, well within acceptable ranges (|skewness|< 2, |kurtosis|< 7), supporting the use of parametric tests.

Results

Baseline equivalence checks

To examine the effectiveness of random assignment, independent samples t-tests were conducted to compare the experimental and control groups on all pre-test measures. As shown in Table 8, no significant differences were found between the two groups on pre-test psychological adaptation (t = − 0.68, p = 0.498), pre-test emotion regulation (t = 0.24, p = 0.812), or pre-test meaning in life (t = 0.14, p = 0.887), with all p-values exceeding 0.05. The effect sizes, as indexed by Cohen’s d, ranged from 0.03 to 0.16 in absolute value, indicating small effects. These results confirm that random assignment was successful, establishing baseline comparability between the two groups prior to the intervention. Nevertheless, to ensure methodological rigor, all subsequent mediation analyses include pre-test scores as covariates.

Table 8.

Baseline comparisons of pre-test scores between groups

Variable	Experimental Group (M ± SD)	Control Group (M ± SD)	t	p	Cohen’s d
PA-pre	46.59 ± 5.27	45.70 ± 5.98	−0.68	0.498	0.16
ER-pre	56.57 ± 3.93	56.78 ± 3.84	0.24	0.812	0.05
MIL-pre	43.84 ± 6.59	44.03 ± 4.69	0.14	0.887	0.03

Cohen’s d was calculated based on pooled standard deviations. PA-pre = Pre-test psychological adaptation, ER-pre = Pre-test emotion regulation, MIL-pre = Pre-test meaning in life. The same abbreviations apply to the following tables

Manipulation check: effect of the intervention on psychological adaptation

To examine whether the intervention successfully enhanced psychological adaptation, a one-way analysis of covariance (ANCOVA) was conducted with post-test psychological adaptation as the dependent variable, pre-test psychological adaptation as the covariate, and group as the fixed factor.

As shown in Table 9, after controlling for pre-test scores, the main effect of group was significant, F = 38.92, p < 0.001, partial η² = 0.35, indicating a large effect. Examination of estimated marginal means (Table 10) revealed that the experimental group reported significantly higher post-test psychological adaptation (M _adj = 52.88, SE = 0.66) compared to the control group (M _adj = 47.01, SE = 0.66), with a mean difference of 5.87 (95% CI [3.99, 7.75], p < 0.001). These results confirm that the adaptation-focused writing intervention effectively improved psychological adaptation, supporting Hypothesis 5.

Table 9.

Manipulation check: ANCOVA for post-test psychological adaptation

Source	F	df	p	Partial η2
PA-pre (covariate)	55.83	1, 71	< 0.001	0.44
Group	38.92	1, 71	< 0.001	0.35

Corrected model R² = 0.59, adjusted R² = 0.58. PA-pre = pre-test psychological adaptation. The same abbreviation apply to the following tables

Table 10.

Adjusted means for post-test psychological adaptation by group

Group	Adjusted Mean	SE	95% CI
Control	47.01	0.66	[45.69, 48.34]
Experimental	52.88	0.66	[51.56, 54.21]

Adjusted means are evaluated at the covariate value PA-pre = 46.15. The mean difference between groups was 5.87, p < 0.001, 95% CI [3.99, 7.75]

Main effects: effects of the intervention on emotion regulation and meaning in life

To examine the impact of the intervention on emotion regulation and meaning in life, two separate one-way analyses of covariance (ANCOVA) were conducted. For emotion regulation, post-test scores were entered as the dependent variable, with pre-test emotion regulation as the covariate and group as the fixed factor. For meaning in life, post-test scores were entered as the dependent variable, with pre-test meaning in life as the covariate and group as the fixed factor.

As shown in Table 11, after controlling for pre-test scores, the main effect of group was significant for both outcomes. For emotion regulation, F = 14.37, p < 0.001, partial η² = 0.17, indicating a medium-to-large effect. For meaning in life, F = 40.25, p < 0.001, partial η² = 0.36, indicating a large effect.

Table 11.

Main effects: ANCOVAs for post-test emotion regulation and meaning in life

Dependent Variable	Source	F	df	p	Partial η2
ER-post	ER-pre (covariate)	54.53	1, 71	< 0.001	0.43
	Group	14.37	1, 71	< 0.001	0.17
MIL-post	MIL-pre (covariate)	61.63	1, 71	< 0.001	0.47
	Group	40.25	1, 71	< 0.001	0.36

For ER-post, model R² = 0.49, adjusted R² = 0.47. For MIL-post, model R² = 0.59, adjusted R² = 0.57. ER-pre = pre-test emotion regulation; MIL-pre = pre-test meaning in life. The same abbreviations apply to the following tables

Examination of estimated marginal means (Table 12) revealed that the experimental group reported significantly higher post-test emotion regulation (M _adj = 59.10, SE = 0.49) compared to the control group (M _adj = 56.47, SE = 0.49), with a mean difference of 2.63 (95% CI [1.25, 4.02], p < 0.001). Similarly, the experimental group reported significantly higher post-test meaning in life (M _adj = 49.90, SE = 0.63) compared to the control group (M _adj = 44.24, SE = 0.63), with a mean difference of 5.66 (95% CI [3.88, 7.44], p < 0.001).

Table 12.

Adjusted means for post-test emotion regulation and meaning in life by group

Dependent Variable	Group	M adj	SE	95% CI	Mean Diff	p
ER-post	Control	56.47	0.49	[55.49, 57.45]	2.63	< 0.001
	Experimental	59.10	0.49	[58.12, 60.08]
MIL-post	Control	44.24	0.63	[42.98, 45.50]	5.66	< 0.001
	Experimental	49.90	0.63	[48.64, 51.15]

M _adj are evaluated at the covariate values: ER-pre = 56.68, MIL-pre = 43.93

Mediation analysis

Mediation analysis was conducted using PROCESS macro Model 6 [32] with 5,000 bootstrap samples to examine the serial mediating roles of psychological adaptation and emotion regulation in the relationship between the intervention and meaning in life. The model specified group as the independent variable, post-test psychological adaptation and post-test emotion regulation as sequential mediators, and post-test meaning in life as the dependent variable, controlling for pre-test psychological adaptation, pre-test emotion regulation, pre-test meaning in life, gender, and age.

Path analysis (Table 13) revealed that the intervention significantly and positively predicted post-test psychological adaptation (B = 5.60, p < 0.001), and post-test psychological adaptation significantly and positively predicted post-test emotion regulation (B = 0.28, p = 0.001). However, post-test emotion regulation did not significantly predict post-test meaning in life (B = 0.02, p = 0.893). The direct effect of post-test psychological adaptation on post-test meaning in life was marginally significant (B = 0.24, p = 0.054). Both the direct effect (B = 4.31, p < 0.001) and total effect (B = 5.72, p < 0.001) of the intervention on post-test meaning in life were significant.

Table 13.

Path coefficients of the serial mediation model

Path	B	SE	t	p	95% CI
Group → PA-post	5.60	0.97	5.75	< 0.001	[3.65, 7.54]
PA-post → ER-post	0.28	0.08	3.35	0.001	[0.11, 0.44]
ER-post → MIL-post	0.02	0.17	0.13	0.893	[−0.32, 0.36]
PA-post → MIL-post	0.24	0.12	1.96	0.054	[−0.00, 0.49]
Direct effect (Group → MIL-post)	4.31	1.12	3.86	< 0.001	[2.08, 6.54]
Total effect (Group → MIL-post)	5.72	0.93	6.18	< 0.001	[3.87, 7.57]

The model controlled for PA-pre, ER-pre, MIL-pre, gender, and age. B = unstandardized coefficient

Indirect effects analysis (Table 14) showed that the total indirect effect was not significant, as its bootstrap confidence interval contained zero (Effect = 1.41, 95% CI [−0.01, 2.81]). Among the specific indirect paths, only the path through post-test psychological adaptation alone (Group → post-test psychological adaptation → post-test meaning in life) was significant (Effect = 1.36, 95% CI [0.14, 3.08]). The path through post-test emotion regulation alone (Group → post-test emotion regulation → post-test meaning in life) and the serial mediation path (Group → post-test psychological adaptation → post-test emotion regulation → post-test meaning in life) were not significant, as their confidence intervals included zero.

Table 14.

Bootstrap analysis of indirect effects

Path	Effect	Boot SE	BootLLCI	BootULCI
Total indirect effect	1.41	0.70	−0.01	2.81
Ind1: Group → PA-post → MIL-post	1.36	0.74	0.14	3.08
Ind2: Group → ER-post → MIL-post	0.02	0.20	−0.46	0.44
Ind3: Group → PA-post → ER-post → MIL-post	0.04	0.29	−0.69	0.52

These results indicate that the intervention enhanced meaning in life primarily through the single mediating pathway of psychological adaptation. Neither the single mediation via emotion regulation nor the serial mediation pathway was significant. Thus, Hypothesis 7 was partially supported (only the psychological adaptation mediation path was significant).

Supplementary analysis: gender moderation

To examine whether the intervention effect on meaning in life was moderated by gender, an additional ANCOVA was conducted with MIL-post as the dependent variable, MIL-pre as the covariate, and group, gender, and their interaction as fixed factors. The Group × Gender interaction was not significant, F = 0.21, p = 0.648, partial η² = 0.003, indicating that the intervention effect did not differ between male and female participants. However, given the limited number of male participants (n = 11) and the very small effect size (ηₚ² = 0.003), a post-hoc power analysis indicated that N = 74 provided only 5% power to detect such a minor interaction. Therefore, this null result should be interpreted with caution, and the absence of gender differences cannot be confirmed without a larger, gender-balanced sample.

Brief discussion

Study 2 employed a randomized controlled experiment to test the causal role of psychological adaptation within the chain model identified in Study 1. A post-hoc power analysis using G-Power 3.1 (Faul et al., 2009) [25] indicated that with N = 74 and five predictors, the study had 80% power to detect an effect size of f ² ≥ 0.23. This was sufficient for the large effects observed on psychological adaptation (ηₚ² = 0.35) and meaning in life (ηₚ² = 0.36) but insufficient for the smaller serial indirect effect, consistent with its non-significance in the mediation model.

The results showed that the adaptation-focused writing intervention successfully enhanced psychological adaptation, emotion regulation, and meaning in life. Mediation analysis revealed that the intervention’s effect on meaning in life was primarily transmitted through psychological adaptation, whereas the serial mediation pathway (psychological adaptation → emotion regulation) was not significant. These findings provide experimental evidence supporting the causal relevance of psychological adaptation as a key mechanism linking the antecedent to meaning in life, consistent with Study 1. However, the failure to replicate the serial mediation effect in the experimental context suggests that the temporal dynamics of this chain may require further investigation, particularly with longer-term follow-ups or more intensive interventions.

Discussion

Amid the multifaceted transition from secondary education to university, the construction and maintenance of meaning in life emerge as a pivotal psychological resource for incoming freshmen. Building on the well-established literature linking adverse childhood experiences to compromised mental health [35], this study delineates a specific pathway through which childhood trauma relates to a key positive psychology outcome during this sensitive developmental period. Three core findings underpin the theoretical and empirical contributions of this study. First, childhood trauma (CT) was associated with lower meaning in life (MIL) both directly and indirectly, aligning with its theorized role in undermining global meaning systems [55]. Second, psychological adaptation (PA) and emotion regulation (ER) each served as significant mediating mechanisms, consistent with models that prioritize adaptive functioning and emotional regulatory skills in the process of meaning-making [30, 55]. Third, and central to our model, it is plausible that these two distinct processes operate in a sequential chain. Specifically, a deficit in the macro-level, context-shaping capacity of PA might give rise to conditions that place demands on the micro-level process of ER, which in turn could be associated with reduced MIL. Notably, PA and ER are process-oriented constructs that describe how individuals negotiate environmental and affective demands, whereas MIL represents an experiential outcome—a subjective evaluation of whether one’s life has purpose and coherence. Decomposing MIL into MLQ-P and MLQ-S revealed a striking asymmetry: the chain model was substantially more powerful in explaining the presence of meaning than the search for meaning, and PA dominated the indirect pathways for MLQ-P whereas ER played a more prominent role for MLQ-S. This pattern offers an extension of the developmental systems view by outlining a potential functional and temporal hierarchy. To address the causal ambiguity inherent in cross-sectional designs, Study 2 employed a randomized controlled experiment to test the causal role of psychological adaptation within this chain model. The experimental results provided causal evidence supporting psychological adaptation as a key mechanism, while also revealing nuanced differences in the chain pathway that inform future research.

Childhood trauma and meaning in life: linking early risk to positive development

Our results confirm a significant negative association between childhood trauma and meaning in life among Chinese university freshmen, supporting Hypothesis 1. This extends the adverse childhood experiences (ACEs) gradient beyond internalizing symptoms to a core positive psychology construct during a targeted life transition [9, 51]. The persistence of a significant direct effect even after accounting for mediators suggests that childhood trauma may impair meaning in life through pathways beyond the broad adaptive and regulatory processes measured here [40, 43, 44]. The meaning-making model provides a coherent explanation: traumatic experiences often shatter fundamental assumptions about the self as worthy and the world as orderly and benevolent [34, 60]. Integrating such violations into a coherent life narrative is a demanding cognitive-emotional task. Without successful integration, a sense of existential confusion, emptiness, or diminished purpose may persist [16, 17, 55]. For these freshmen actively engaged in identity reconstruction, these unresolved negative schemas may be particularly accessible, directly contaminating efforts to find purpose and value in their new academic and social pursuits [49]. Importantly, the association with MLQ-S was substantially weaker than that with MLQ-P, consistent with the view that the search for meaning may be driven partly by normative developmental forces less easily disrupted by early adversity [6]. This also leaves room for the possibility that some trauma-affected freshmen retain the motivational drive to seek meaning, a process that may, under supportive conditions, facilitate post-traumatic growth [38]. From this perspective, the chain mediation identified in our model—from psychological adaptation to emotion regulation to meaning—may represent not only a pathway of vulnerability but also a potential pathway of recovery: interventions that strengthen adaptive and regulatory capacities could help transform the lingering cognitive-emotional impact of traumatic memories into a renewed search for coherence and purpose [66, 69]. Furthermore, these findings underscore that the freshman transition itself—as a period of identity reconstruction and heightened exploration—may moderate how the search for meaning is expressed and should be interpreted. What appears as elevated searching in this population may reflect, at least in part, the normative developmental demands of emerging adulthood rather than the psychological consequences of early adversity alone [6]. More broadly, recent experimental work illustrates that trauma-related memories do not operate merely as static distal risk factors, but actively interact with intervening psychological processes to shape later outcomes [12].

Psychological adaptation as a conduit from trauma to meaning

The significant mediating role of psychological adaptation, as posited in Hypothesis 2, underscores its function as a primary mechanism. Childhood trauma, by disrupting the development of trust, safety, and effective coping, can leave individuals ill-equipped to meet the demands of novel environments [57, 62]. For Chinese university freshmen, entering university with compromised adaptive resources likely means facing academic pressures, social complexities, and autonomy-related challenges with a heightened sense of overwhelm and poor environmental mastery [4, 14]. This state of maladaptation directly drains the psychological resources necessary for engaging in exploratory behaviors, committing to new goals, and reflecting on life’s direction—processes fundamental to the construction of meaning [24, 55]. Our finding that this path accounted for the largest proportion of the indirect effect—particularly for MLQ-P—highlights psychological adaptation as a critical leverage point for interventions aiming to bolster the experience of meaning among trauma-affected Chinese university freshmen.

Study 2 provided experimental evidence for this causal interpretation. The adaptation-focused writing intervention successfully enhanced psychological adaptation (manipulation check), and this improvement in turn predicted increased meaning in life. Moreover, the indirect effect of the intervention on meaning in life through psychological adaptation was significant, whereas the pathway through emotion regulation alone was not. These experimental findings corroborate the cross-sectional results from Study 1 and provide causal evidence consistent with psychological adaptation shaping meaning in life. Taken together, both studies converge on psychological adaptation as the dominant mechanism linking distal antecedents (whether childhood trauma or experimental manipulation) to meaning in life. We acknowledge, however, that the intervention likely engages positive reflection and self-affirmation alongside adaptation per se, and future research with more targeted comparison conditions is needed to isolate the specific contribution of each component.

Emotion regulation as a proximal mediator in meaning-making

The confirmation of Hypothesis 3 reinforces the centrality of emotion regulation in the meaning-making process [30, 45]. Childhood trauma is a potent disruptor of emotion regulation development, often fostering reliance on maladaptive strategies like emotional suppression or rumination [5]. For these freshmen, the transition itself is an emotion-laden stressor. Those entering with pre-existing regulatory deficits may experience more intense and prolonged negative affect in response to academic setbacks or social friction. Emotion dysregulation consumes attentional and cognitive resources that could otherwise be directed toward integrative meaning-making processes, such as reflecting on experiences or forging future plans [47]. Consequently, the inability to effectively manage emotional responses can act as a proximal barrier, preventing individuals from perceiving coherence, purpose, and significance in their daily university life and broader future trajectory [3, 59]. Crucially, this micro-level failure to manage immediate emotional responses is conceptually distinct from the macro-level failure to adapt to the university context. Our model and findings suggest that the latter often sets the stage for the former. The modest correlation (r = 0.16) between the two constructs in Study 1 underscores that while related, they are empirically distinguishable. Further supporting this distinction, ER was a considerably stronger predictor of MLQ-S than of MLQ-P, suggesting that the active pursuit of meaning relies more heavily on the capacity to navigate emotional experiences than does the relatively stable experience of having meaning.

In Study 2, the intervention also significantly improved emotion regulation, consistent with the idea that enhancing psychological adaptation may have downstream benefits for regulatory capacity [70]. However, the indirect effect of the intervention on meaning in life through emotion regulation alone was not significant, and emotion regulation did not significantly predict meaning in life in the mediation model. This suggests that while emotion regulation is correlated with meaning in life and responsive to the intervention, it may not function as an independent causal mediator within the short timeframe of the experiment. The temporal dynamics of emotion regulation in the meaning-making process warrant further investigation. Future research could also draw on dual-process frameworks (e.g., coping flexibility; [19]) to specify how ER strategies differentially operate across the loss-oriented and restoration-oriented processes that characterize trauma survivors’ attempts to construct coherent life narratives.

The chain mediating pathway: a hierarchical cascade of impairment and its experimental test

The support for Hypothesis 4 provides the most nuanced insight, revealing a sequential process rather than two parallel mediation paths. While modest in magnitude, this chain effect is theoretically meaningful for understanding the mechanism linking childhood trauma to reduced meaning in life. It suggests that the impact of childhood trauma on meaning in life is not merely additive but follows a logical functional hierarchy: early adversity first erodes individuals’ broad capacity to achieve person-environment fit (psychological adaptation). This compromised adaptation, characterized by persistent stress and poor adjustment to university life, then creates a context that chronically taxes and overwhelms the more specific system responsible for managing emotional responses (emotion regulation) [2, 42]. Finally, this resultant emotion dysregulation directly interferes with the cognitive and affective processes required for sustaining meaning in life [55]. This pattern aligns with a developmental psychopathology framework where deficits in higher-order adaptive systems are hypothesized to cascade to disrupt more specific regulatory processes, which in turn impact discrete outcomes like meaning [50].

Study 2 tested this causal cascade by experimentally manipulating psychological adaptation. Contrary to expectations, the chain mediation pathway (intervention → psychological adaptation → emotion regulation → meaning in life) did not reach statistical significance, in contrast to the cross-sectional findings from Study 1. Several explanations may account for this discrepancy. First, the temporal scale of the experiment may have been insufficient for the cascade to unfold. The chain model implies a hypothesized sequential process that may require more time to propagate from broader adaptation to specific regulatory skills and ultimately to meaning. The immediate post-test assessment may have captured the direct effects on psychological adaptation but missed the downstream effects on emotion regulation that could emerge over days or weeks. Second, the single-session writing intervention, while effective in shifting psychological adaptation, may not have been intensive enough to trigger detectable change in the more habitual process of emotion regulation. Third, the sample size (N = 74), while adequately powered for detecting medium-to-large effects, may have been insufficient to detect the small serial indirect effect (post-hoc sensitivity analysis confirmed 80% power only for f ² ≥ 0.23, whereas the serial indirect effect fell below this threshold). Fourth, the association between PA and ER may be reciprocal rather than strictly unidirectional. Although the present study only tested the hypothesized sequence, future research employing cross-lagged panel designs could formally compare alternative orderings and reciprocal causal models. This interpretation is consistent with the cross-sectional findings from Study 1, where the modest empirical overlap between PA and ER (r = 0.16) suggests they function as sequential stages in a developmental cascade rather than as interchangeable indicators of a single maladjustment process. These explanations are not mutually exclusive and suggest that the absence of the chain pathway in the experimental context does not invalidate the theoretical model but rather highlights the need for future research to examine the temporal dynamics of these processes using longer-term follow-ups and more intensive interventions.

Importantly, the absence of the chain pathway in the experimental context does not indicate that the chain model is invalid, but rather suggests that its manifestation may depend on the temporal scale of observation. Cross-sectional designs capture the accumulated outcome of ongoing processes, whereas experimental designs with immediate post-test assessment capture only the initial stage of the cascade. Future studies employing longitudinal follow-ups (e.g., one week, one month) after an intervention are needed to test whether the chain effect emerges over time.

Practical implications for university support systems

The combined findings from both studies offer several implications for practice. First, routine assessment of psychological adaptation during the freshman transition could help identify those freshmen who are struggling to adjust, which our findings suggest is a proximal risk factor for diminished meaning in life, particularly the presence of meaning. Early identification allows for timely intervention before maladaptive patterns become entrenched. Second, interventions should prioritize psychological adaptation as a primary target. Study 2 demonstrated that a brief adaptation-focused writing intervention—prompting students to reflect on challenges, identify coping experiences, and formulate future strategies—can effectively enhance psychological adaptation and, through this mechanism, improve meaning in life. Such low-cost, scalable interventions can be integrated into university orientation programs or first-year experience courses. Third, while emotion regulation did not emerge as an independent causal mediator in the experimental context, its correlation with meaning in life suggests it remains a valuable secondary target [30, 45]. Universities may consider integrating training in adaptive emotion regulation strategies (e.g., cognitive reappraisal, mindfulness) as a complement to adaptation-focused efforts. For students whose adaptation difficulties appear entrenched or resistant to skills-based approaches, deeper therapeutic work addressing underlying schemas—including those potentially rooted in early adversity—may be warranted, though this goes beyond the scope of what university-wide programs can provide.

Limitations and future directions

Several limitations should be considered when interpreting these findings, and they point to directions for future research.

First, Study 1 employed a cross-sectional design, which precludes causal or directional conclusions about the proposed chain. Although Study 2 addressed this limitation by experimentally manipulating psychological adaptation, the chain mediation pathway was not replicated in the immediate post-test assessment. As discussed in "The chain mediating pathway: a hierarchical cascade of impairment and its experimental test" section, this may reflect both the temporal scale of the cascade and the small sample size (N = 74), which a post-hoc power analysis confirmed was underpowered for detecting small indirect effects (f ² ≥ 0.23 required for 80% power). Future research should employ longitudinal designs with multiple follow-up assessments and larger samples to examine whether the chain effect unfolds over time.

Second, both studies relied on self-report measures, which may inflate associations due to common method variance. Additionally, our use of the ERQ total score combined cognitive reappraisal and expressive suppression—two strategies that often show opposing relationships with well-being [29]—which may have attenuated some emotion-regulation-related indirect effects. Although preliminary checks (Harman’s single-factor test) suggested this was not severe, future research should incorporate multi-method assessments (e.g., behavioral tasks for emotion regulation, peer or instructor reports for adaptation).

Third, while we controlled for basic demographics including SES, other potential confounding variables (e.g., concurrent depression, anxiety, or personality traits) were not included. Future models should test the robustness of this chain pathway against these covariates.

Fourth, Study 2 used the MLQ total score as the outcome variable rather than preregistering separate analyses for MLQ-P and MLQ-S. Based on the cross-sectional findings from Study 1, which showed that psychological adaptation primarily predicts MLQ-P rather than MLQ-S, it is plausible that the intervention’s significant effect on the total MLQ score was driven largely by improvements in the presence of meaning. Future experimental research should include both MLQ subscales as distinct outcome measures to examine whether adaptation-focused interventions differentially affect the experience of meaning versus the active search for it.

Fifth, although this study adopted a PTG-informed perspective in the discussion section, it did not directly measure post-traumatic growth, nor did it examine how specific maltreatment subtypes may differentially affect the proposed pathways [68]. Future research could explicitly integrate PTG measures and subtype-specific analyses to test whether the chain mediation pathway operates differentially for those who report growth versus those who do not, thereby directly examining the trauma memory–regulation–meaning linkage proposed in contemporary PTG frameworks [67].

Finally, the measure of psychological adaptation, though validated in Chinese samples, is culturally embedded. Its operationalization and transferability to other cultural contexts warrant examination. Future work should also explore additional mediators (e.g., self-esteem, social support) and moderators (e.g., resilience, campus belonging) to paint a more complete picture of how childhood trauma influences the construction of meaning in life during educational transitions. Moreover, both studies were conducted among Chinese university students and Study 2 exhibited an unbalanced gender distribution. These limitations warrant future replication across more culturally diverse backgrounds and gender-balanced samples.

Conclusion

This study clarifies the association between childhood trauma and meaning in life among university freshmen by testing a self-regulatory chain mediating model. Cross-sectional findings from Study 1 revealed that childhood trauma was associated with lower meaning in life both directly and indirectly, primarily through psychological adaptation. This mechanism operated predominantly through the presence of meaning, whereas the search for meaning showed a weaker and more distributed pattern of indirect effects. A chain mediating effect through psychological adaptation and emotion regulation in sequence was observed for both dimensions, though the model explained substantially more variance in MLQ-P than in MLQ-S. Experimental findings from Study 2 provided causal evidence that enhancing psychological adaptation improves meaning in life, supporting its role as a key mechanism. However, the chain mediation pathway was not replicated in the experimental context, highlighting the need for future research to examine the temporal dynamics of these processes. Taken together, this study advances a process-oriented understanding of the relationship between childhood trauma and meaning in life from a positive developmental perspective, suggesting that the active search for meaning may represent a distinct process that is not simply the inverse of diminished presence, and offering both theoretical insights and practical guidance for fostering existential well-being among university freshmen.

Authors’ contributions

GQZ was responsible for the overall study design, data collection for both Study 1 and Study 2, interpretation of data, manuscript writing, and revision. MHL contributed to data collection, data analysis, and served as a key executor of the experimental intervention in Study 2. LGY contributed to the editing and polishing of the manuscript. MHL, HS, and XJH contributed to the revision of the manuscript. ZLZ contributed to the supervision of the manuscript. All authors contributed to the article and approved the submitted version.

Funding

This study was supported by grants from the Sichuan Higher Education Talent Development Quality and Teaching Reform Project (JG2410008); the Sichuan Academic Achievement Analysis and Application Research Center Project (SCAA24-B16); and the Sichuan Educational Informatization Application and Development Research Center Project (JYXX2428).

Data availability

The datasets used and analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

This study was conducted in accordance with the principles outlined in the Declaration of Helsinki. The research protocol underwent formal review and received approval from the Ethics Committee of Xihua University (approval number XHU-260403). Written informed consent was obtained from all participants for their participation in this study.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.