ABSTRACT
The transition to motherhood is a period of psychological vulnerability but also offers important opportunities to refine and strengthen resilience capacities that support long‐term mental health. The Systematic Self‐Reflection Model of Resilience Strengthening proposes dynamic, reciprocal relationships between self‐reflection, coping insight, and capacities for resilience, but this model has not been tested during the transition to motherhood. We aimed to clarify: (i) how self‐reflection and coping insight change across this life transition, and (ii) to examine the relationships among self‐reflection, coping insight, and depressive symptoms in first‐time mothers. It explored their bidirectional and indirect interactions, with a focus on how these processes contribute to resilience and mental health. First‐time pregnant women in Australia (N = 147) completed longitudinal surveys across three time points (two prenatal, one postnatal). Multi‐level modeling assessed changes in self‐reflection, coping insight, and depressive symptoms over time, and tested mediating relationships at both within‐ and between‐person levels. Cross‐lagged panel models evaluated the directionality and reciprocity of these associations. Findings showed that coping insight consistently predicted lower depressive symptoms across multiple models. However, the pathway from self‐reflection to insight was less stable, suggesting that reflection alone may not reliably foster adaptive insight. Conversely, depressive symptoms more consistently influenced later self‐reflection indirectly via reduced insight, indicating that depression may undermine cognitive processes that support resilience. These findings provide novel evidence of how self‐reflection, coping insight, and depression interact during the transition to motherhood, highlighting the importance of targeting coping insight in perinatal interventions to strengthen resilience and protect maternal mental health.
Keywords: coping, depression, maternal mental health, meaning‐making, pregnancy, resilience
1. Introduction
Motherhood is a major life transition. First‐time pregnancy brings profound physical changes, the formation of maternal identity, and the anticipation of caring for a new life; an experience that is both challenging and transformative (McMahon 2023). Mental health challenges are common during pregnancy and early parenting, making mental illness the most prevalent morbidity associated with childbearing (Howard and Khalifeh 2020). A comprehensive meta‐analysis of 565 studies across 80 countries found that postpartum depression affects approximately 17.22% of women worldwide, indicating that nearly one in five women experience clinically significant depressive symptoms after childbirth (Wang et al. 2021). Pregnant women experience mental health problems at rates two to three times higher than non‐pregnant women (E. P. Davis and Narayan 2020). Given the well‐documented negative impact of perinatal depression on maternal well‐being and child development (Stein et al. 2014), there is growing interest in the cognitive and emotional processes that support psychological adaptation during this life transition.
The Systematic Self‐reflection Model of Resilience Strengthening (Crane, Boga, et al. 2019), proposes that coping self‐reflection and related insights are metacognitive processes that enhance resilience during transitional life phases, such as first‐time motherhood (Crane, Boga, et al. 2019; Falon et al. 2021). Coping self‐reflection is a practice whereby individuals examine their emotions, cognitions, and actions during coping, which is argued to stimulate the emergence of coping‐related insights that guide growth, learning, and resource development from stressful experiences (Crane, Boga, et al. 2019; Falon et al. 2021). Previous work suggests that transitional periods, such as first‐time motherhood, may be a key period for self‐reflection (Heine et al. 2025). Research shows that destabilizing life events (e.g., moving, job change) are linked to a stronger desire for accurate self‐knowledge, indicating that personal upheaval may prompt greater self‐reflection. Other work considers metacognitive processes to be relatively stable and trait like (Faissner et al. 2018; Silvia and Phillips 2011). However, the change in self‐reflection and insight and the precise relationship between self‐reflection, insight, and mental ill‐health symptoms during major life transitions remains unclear.
Previous theories propose a unidirectional pathway in which self‐reflection fosters insight, supporting emotional regulation and reducing depressive symptoms (Dimaggio et al. 2009; García‐Mieres et al. 2020; Grant et al. 2002). These accounts suggest that self‐reflection helps individuals monitor and assess their experiences, and that insight clarifies this information to guide goal‐directed behavioral adjustments that promote mental health (Grant et al. 2002). However, evidence indicates a more complex pattern, whereby depressive symptoms potentially hinder insight and, consequently, adaptive forms of reflective thinking (Silvia and Phillips 2011). Additionally, some studies suggest that self‐reflection and insight are uncorrelated (Grant et al. 2002; Silvia and Phillips 2011) or that self‐reflection has a mixed association with mental health outcomes (He and Gan 2025). Insight, however, consistently emerges as a key protective factor for depression and anxiety (Grant et al. 2002; Silvia and Phillips 2011). This study investigated changes in coping self‐reflection and coping insight across the transition to motherhood and their associations with depressive symptoms in first‐time mothers. We examined bidirectional and indirect relationships among these processes, with an emphasis on their role in resilience and mental health.
1.1. The Significance of Perinatal Mental Ill‐Health
All women face an increased risk of mental health disorders during the perinatal period, with perinatal depression being one of the most common (Brockington 2004; Gavin et al. 2005). While primigravida mothers are experiencing the novel physiological, psychological, and social changes of pregnancy for the first time, which can serve as a stressor (Alves et al. 2021), rates of perinatal depression are similar for women in their first and subsequent pregnancies (e.g., Bassi et al. 2017; Zhang et al. 2024), but first‐time mothers demonstrate a higher risk of requiring psychiatric admission than mothers giving birth to their second or subsequent child (Munk‐Olsen et al. 2006). Therefore, the perinatal period broadly presents a risk for the mental health of the mother. Research demonstrates that depressive symptoms in pregnancy and postpartum women do not follow a single, uniform course but are highly variable. While most women experience consistently low or moderate symptoms, a small subset show notable change over time, either increasing after birth, or beginning pregnancy with high symptoms that decline postpartum before rising again in later years. This highlights that symptoms can emerge, persist, or re‐emerge well beyond the early postpartum period, underscoring the importance of ongoing mental health monitoring across the transition to motherhood (Ahmed et al. 2019).
Maternal mental health issues during the perinatal period not only affect mothers but also pose risks to child development. While developmental disturbances are not inevitable, they can have lasting effects on individuals and society. In Australia, maternal mental health difficulties are estimated to cost $877 million annually, with a lifetime economic impact of $5.2 billion (PwC Consulting Australia 2019), while in the UK, around 60% of the total estimated economic burden reflects impacts on children (Bauer et al. 2016). Addressing maternal mental health is therefore essential to reducing both personal and societal costs and promoting healthier developmental outcomes across generations.
2. Resilience to Depressive Symptoms During the Perinatal Period
While reducing risk is essential for promoting maternal mental health, fostering resilience is equally important. This requires a clearer understanding of the mechanisms that enable women to adapt during the transition to motherhood. Although resilience has been widely studied, its development in new parents remains poorly understood (Young and Ayers 2021). The current study focuses on resilience to the onset of perinatal depressive symptoms. Drawing on established definitions, resilience in this context refers to a woman's capacity to adapt to the stresses of pregnancy and navigate this period with minimal or no symptoms of depression (Kalisch et al. 2017).
Studies suggest several features likely to support resilience to depressive symptoms during the perinatal period. These include: strong support networks, including partners, family, friends, and healthcare providers (Hu et al. 2019); physical exercise, particularly in non‐depressed women (Daley et al. 2015); yoga (Gong et al. 2015); and positive early‐life experiences (e.g., supportive childhood relationships) (E. P. Davis and Narayan 2020; Yehuda and Meaney 2018). The longitudinal Growing Up in New Zealand study identified a range of resources associated with resilience across the perinatal period, broadly categorized as internal (e.g., physical health, attitudes to parenthood) and external (e.g., social support, sense of community belonging; Farewell et al. 2022).
The COVID‐19 pandemic prompted research on resilience during the transition to motherhood, highlighting how pregnant women coped with this substantial external stressor. Consistent with broader resilience frameworks, studies identified protective influences across personal (e.g., active coping, meaning making), contextual (e.g., financial stability, childcare access), and relational domains (e.g., social and partner support; Thompson et al. 2022). A multisystemic perspective (Young and Ayers 2021) further underscored the interplay of individual attributes, relationships, support systems, cultural context, and environmental conditions. Collectively, this work reinforces that resilience in the perinatal period is multifaceted and depends on both internal and external resources.
Despite growing research on resilience, how new mothers actively build and refine it remains underexplored. While the literature highlights the types of resilience capacities that support maternal well‐being, it offers far less insight into how these capacities develop. The processes by which individuals build or strengthen these internal resources remain unclear, and it is these developmental mechanisms that the present study aims to illuminate. Our focus is on the psychological processes that may underpin the development and strengthening of both internal and external resilience factors. For example, individuals do not simply possess resources; many actively create, access, or sustain them. The availability and usefulness of external supports are shaped not only by environmental conditions but also by an individual's willingness and capacity to engage with those supports. Moreover, some people may have greater access to resources because of characteristics, such as higher emotional intelligence, which can influence how they recognize, seek, and utilize support (e.g., Ciarrochi et al. 2003). Adaptation to motherhood is not passive. While some resilience factors, such as family background, are fixed, many can be strengthened through modifiable internal and external assets (e.g., social engagement, building support networks, healthy behaviors, and refining coping strategies).
We propose that adaptation supporting resilience to depression during first‐time motherhood is an interactive process, in which mothers explore, evaluate, and select supports that best meet their unique needs. This tailored approach aligns with the coping flexibility perspective (Kato 2012), which highlights the importance of discontinuing ineffective coping strategies and adopting alternative ones. Effective adaptation requires continually aligning coping strategies with emerging stressors through an ongoing evaluation–adjustment cycle. Evidence highlights a growing recognition of resilience as a protective factor in perinatal mental health, yet current intervention models remain predominantly reactive rather than preventive (Hajure et al. 2024).
2.1. The Development of Capacities for Resilience During the Peri‐ and Postnatal Period: The Role of Coping Self‐Reflection and Coping‐Related Insight
Although first‐time motherhood carries a heightened risk of depressive symptoms, it also offers a unique opportunity to develop capacities for resilience. The novel stressors, setbacks, and disrupted routines of this transition can prompt a shift from automatic coping to more reflective processes that refine adaptive functioning. Drawing on resilience metatheories (Richardson 2002), periods of disruption are conceptualized as developmental opportunities through which individuals can reintegrate at higher levels of functioning. The perinatal transition may therefore serve as a natural context for resilience capacity development, provided mothers engage in reflective processes that foster insight and adaptive adjustment. However, the application of resilience capacity development theories to this transition remains in its early stages (Hannon et al. 2022; Young and Ayers 2021).
Previous work measuring self‐reflection has demonstrated its negative association with depressive symptoms (Takano and Tanno 2009; Treynor et al. 2003). The protective role of self‐reflection is articulated in the Systematic Self‐Reflection Model of Resilience Strengthening (Crane, Boga, et al. 2019), which proposes that resilience capacities are developed and refined during major life transitions. These resilience capacities are the modifiable psychosocial assets and protective factors that increase the likelihood of resilience (and reduce the likelihood of mental ill‐health). They include: (i) the application of flexible coping strategies, (ii) resilient beliefs, and (iii) coping‐enabling resources. The use of flexible coping strategies refers to context‐sensitive approaches for managing stressors, such as problem solving, positive reappraisal, and emotion regulation, that can be refined over time through reflective learning. Coping‐enabling resources encompass the practical, motivational, and social assets (e.g., supportive relationships) that enable effective coping and can be expanded or strengthened through proactive resource‐seeking and help utilization. Finally, resilient beliefs capture beliefs about agency and control over desired outcomes (e.g., coping self‐efficacy). These capacities function as dynamic and mutually reinforcing systems, each influencing and strengthening the others, thereby increasing the likelihood of successful adaptation and sustained resilience across future stressor experiences (Crane, Boga, et al. 2019).
The transition to new motherhood is marked by heightened stress as familiar routines are disrupted and new demands emerge. These novel stressors may introduce setbacks that require a shift from habitual coping strategies towards reflective thinking. According to the Systematic Self‐Reflection Model, everyday challenges can activate self‐reflective processes that build coping insight, such as recognizing strengths, stress triggers, and available supports. Coping self‐reflection involves the deliberate and conscious practice of examining one's reaction to stressors, stress triggers, considering how stressors contribute to growth, assessing coping effectiveness in relation to values and goals, and identifying future strategies and resources (Crane, Boga, et al. 2019). It has been shown to be associated with self‐insight, increased resilience, general wellbeing, and lower psychological distress (Bucknell et al. 2022; García‐Mieres et al. 2020; Takano and Tanno 2009). Increasing emotional awareness allows individuals to become better at naming emotions and understanding their causes, leading to enhanced emotion regulation (Dimaggio et al. 2009). Adaptive forms of reflection, often studied in coaching and educational contexts, increase self‐awareness, improve coping effectiveness, and support flexible problem‐solving. As a metacognitive skill, self‐reflection helps individuals better understand their thoughts and behaviors, regulate emotions, and refine their coping approaches over time (Chi et al. 1994; Ellis et al. 2014; Gale et al. 2016; Grant et al. 2009). Recent studies show that structured reflective journaling (i.e., the systematic analysis of experiences, successes, and challenges) can foster coping insight, improve emotion regulation, and reduce anxiety and depression during periods of stress (Crane, Boga, et al. 2019; Falon et al. 2021). This work on adaptive coping self‐reflection and its use as an intervention suggests promise for its value in strengthening resilience capacities and reducing vulnerability to mental health difficulties.
The Self‐Reflection and Coping Insight Framework (Falon et al. 2021) distinguishes coping‐related insights from general insight. General insight reflects awareness of one's internal states (Grant et al. 2002), whereas coping‐related insights are understandings of the self, others, and the environment in relation to the stress experience that may support resilience capacity development and refinement (Falon et al. 2021). Coping insights specially support the development of resilience capacities by providing the understanding needed to guide adaptive change. This includes adjusting expectations, priorities, or values in response to new insights, as well as adopting strategies or resources previously lacking. These insights are multidimensional, encompassing awareness of cognitive responses, expectations, and the effectiveness of coping strategies and behaviors, which together shape future adjustments.
Building on narrative and meaning‐making research (McAdams 2001; Park 2010), reflective processes allow individuals to reconstruct stressful experiences into coherent self‐understandings. However, unlike meaning‐making approaches, which focus on coping with the present event by restoring coherence between life experiences and broader cognitive belief systems, the Systematic Self‐Reflection Model extends this view by proposing that insights can also drive functional transformation by refining coping resources and flexible coping strategies for both current and future demands. Within this framework, self‐reflection and coping insight act as metacognitive bridges that translate self‐understanding into adaptive change, increasing the likelihood of resilience in the face of recurring stressors such as those encountered in new motherhood. As women navigate identity shifts, physical changes, emotional demands, and caregiving responsibilities, coping self‐reflection and the insights it generates may play a critical role in their adaptation and well‐being across future parenting challenges.
2.2. The Reciprocal Relationship Between Depressive Symptoms, Insight, and Self‐Reflection
The interplay between reflection, coping insight, and depressive symptoms can be conceptualized within a dynamic self‐regulatory system. Adaptive self‐reflection facilitates coping‐related insights that strengthen regulatory flexibility and resilience, while elevated depressive symptoms may undermine these metacognitive processes through cognitive rigidity, undermined executive functioning, negative bias, and avoidance (R. N. Davis and Nolen‐Hoeksema 2000). Moreover, ruminative thoughts associated with depression may also limit self‐reflection and insight. Individuals who engage in rumination tend to fixate rigidly on negative emotions and personal failures, reinforcing cognitive distortions and emotional distress (Nolen‐Hoeksema 2000). Related research shows that prolonged cortisol elevation associated with depression diminishes cognitive flexibility and decision‐making (Shields et al. 2016), constraining the effectiveness of stress‐regulation and metacognitive strategies that rely on executive function. A meta‐analysis of 113 studies has similarly demonstrated that major depressive disorder is associated with broad impairment in multiple aspects of executive function (Snyder 2013). When executive function is compromised, processes necessary for insight (e.g., regulating attention, evaluating mental states, and integrating emotional experiences), become more difficult. Conversely, lower levels of depressive symptoms may better support the emergence of insight.
In addition, greater levels of insight may feed back into refined capacities for self‐reflection. Drawing from regulatory flexibility models (Aldao et al. 2015; Bonanno and Burton 2013), insight can be viewed as a metacognitive outcome of monitoring and evaluating self‐regulation strategies, which then informs future situational appraisals. In this way, insight functions as information within the self‐regulatory system: understanding why certain strategies work enables more intentional self‐reflection, more accurate interpretation of experiences, and better‐adjusted coping efforts. Over time, increases in insight may therefore strengthen reflective capacity, support more accurate self‐appraisal, and promote more adaptive meaning‐making in response to stress. Previous empirical work supports the fundamental relationships proposed. Coping self‐reflection (but not general reflection) was associated with greater levels of insight, which in turn enhanced perceived resilience and general wellbeing (Bucknell et al. 2022). On the other hand, general insight has been demonstrated to have a non‐significant relationship with self‐reflection (Silvia and Phillips 2011). However, no work has explicitly investigated the potential for mediation and bi‐directional paths. This bidirectional relationship reflects both feedforward and feedback dynamics: while increased coping self‐reflection may lead to greater insight and lower depression, depression may also disrupt the emergence of insight and impair adaptive coping self‐reflection, necessitating interventions that help individuals navigate reflective processes in an adaptive manner.
2.3. The Current Study
Although guided interventions can enhance coping self‐reflection and support mental health, it is unclear whether coping reflection naturally emerges during major life transitions, whether it fosters coping‐related insight, and in turn lowers depressive symptoms. The reciprocal relationships among these processes also remain largely unexplored. This paper sought to advance our understanding of how coping self‐reflection and coping insight interact with depressive symptoms during the transition to motherhood, and how these metacognitive processes contribute to resilience. Using a longitudinal design across three perinatal time points (24–32 weeks' gestation, 37–39 weeks' gestation, and 1‐month postpartum), we tested the Systematic Self‐Reflection Model of Resilience Strengthening in a perinatal population for the first time, to explore whether coping self‐reflection contributes to reductions in depressive symptoms through increases in specific coping insights, and whether this indirect relationship is bidirectional. In doing so, we aimed to clarify the dynamic mechanisms through which self‐reflection and insight contribute to maternal mental health during early motherhood.
RQ 1
Do average levels of self‐reflection, coping insight, and depression change across the transition to motherhood, as first‐time mothers adjust to this major life event?.
H 1
(Multilevel Indirect Effects Hypothesis): There will be an indirect relationship between self‐reflection and depressive symptoms via coping insight at both the within‐person and between‐person levels, with stronger effects expected at the between‐person level. Specifically:
Within‐Person Indirect Effect: When individuals engage in more self‐reflection than usual, they will report greater coping insight, which in turn will be associated with reduced depressive symptoms.
Between‐Person Indirect Effect: Individuals who reflect more on average will show higher overall coping insight, which will be linked to lower average depressive symptoms.
H 2
(Bidirectional Indirect Effects Hypothesis): Cross‐lagged panel models will demonstrate that the best‐fitting models are reciprocal (fully crossed), supporting bidirectional indirect effects among self‐reflection, coping insight, and depressive symptoms. Specifically, increases in self‐reflection will lead to greater coping insight, which in turn will predict reductions in depressive symptoms over time. Conversely, higher depressive symptoms will lead to lower coping insight, which will subsequently reduce self‐reflection.
3. Method
3.1. Data Transparency and Openness
The datasets used in this analysis are available at https://figshare.com/s/351c0276aabb8b417678 and can be downloaded by requesting access from the corresponding author.
3.2. Participants and Design
This longitudinal survey study included three time points: two during pregnancy and one at 1‐month postpartum. Participants were 212 first‐time pregnant women between 24 and 32 weeks' gestation, residing in Australia. Exclusion criteria included: (1) insufficient English to complete study measures, (2) lack of access to required digital technology, (3) previous live birth, (4) gestational age outside the 24–32‐week window at study entry, and (5) discrepancies greater than 5 weeks between reported gestational age and due date. Eight participants were excluded based on these criteria, resulting in a final sample of 204. Of these, 147 participants completed measures at all three time points (see Figure 1 for study flow diagram). See Table 1 for sample descriptives.
FIGURE 1.
Study diagram illustrating participant retention and attrition at each study phase.
TABLE 1.
Participant descriptive statistics based on N = 204.
| Demographic information | M (SD), range |
|---|---|
| Age at enrollment (years) | 32.00 (3.89), 18‐42 |
| Gestation at enrollment (weeks) | 28.12 (2.38), 24‐32 |
| n (%) | |
|---|---|
| Country of birth | |
| Australia | 152 (74.51) |
| Other | 52 (25.49) |
| Home language | |
| English only | 163 (79.90) |
| Other | 41 (20.10) |
| Living situation | |
| With partner/spouse | 189 (92.65) |
| With partner/spouse & others | 9 (4.41) |
| With others | 3 (1.47) |
| Alone | 3 (1.47) |
| Education | |
| University | 165 (80.88) |
| School/Technical | 39 (19.12) |
| Location type | |
| Urban | 158 (77.45) |
| Rural | 44 (21.57) |
| Remote | 2 (0.98) |
3.3. Procedure
Ethics approval was obtained from the Macquarie University Human Research Ethics Committee (Ref: 520251100665857). Data were collected as part of a larger programme of research investigating mental health during the transition to first‐time motherhood. Participants were recruited through social media (92% via Facebook), obstetric and midwifery practices, antenatal clinics, educational and perinatal psychology services, childcare centers, and professional and personal networks, including paid advertisements. Participants who completed all study components could opt into a draw for one of five AUD $200 gift vouchers. Study information and consent were provided via REDCap accessed through a QR code or study link. Participants completed three online surveys: Time 1 (24–32 weeks' gestation, immediately after consent), Time 2 (37–39 weeks' gestation, sent three weeks before the due date with two reminders), and Time 3 (1 month postpartum, sent 30 days after the due date with two reminders). Because mental health symptoms can emerge at different times during and after pregnancy, previous research does not prescribe a uniform measurement schedule (Ahmed et al. 2019). However, multiple assessments are recommended. We therefore selected timepoints that correspond to periods of heightened stress: mid‐pregnancy, when physical and emotional demands begin to intensify, and late pregnancy, when anticipatory stress commonly increases. The 1‐month postpartum assessment captures a period in which some women show symptom escalation (Ahmed et al. 2019). While we would have ideally preferred to capture women 3‐month postpartum aligned with clinical screening recommendations (Highet and Center of Perinatal Excellence 2023), funding restrictions meant that an earlier time point was necessary. Finally, the greater than a 1 month interval between time points was anticipated to allow the observation of symptom change. All surveys were completed on personal digital devices via REDCap. Participant flow and completion rates are shown in Figure 1.
3.4. Measures
Primary outcomes were depressive symptoms, coping self‐reflection, and coping insight. At Time 1, participants also completed 10 general demographic items, including age, state of residence (with urban or rural/remote classification), education level, home language, and gestational age. Internal consistency for all multi‐item outcome scales was assessed using McDonald's Omega (ω), with Satorra‐Bentler (SBr) used for two‐item scales. Confirmatory factor analysis (CFA) model fit statistics for all outcome scales at each time point are reported in Supporting Information S1: Table S1.
3.5. Depressive Symptoms
Depressive symptoms were measured using the Edinburgh Postnatal Depression Scale (EPDS; Cox et al. 1987), the most widely used screening tool for identifying depressive symptoms in perinatal care (Levis et al. 2020). Originally developed for the postnatal period, the EPDS has also demonstrated adequate sensitivity and specificity for use during pregnancy (Levis et al. 2020). The scale consists of 10 items rated on a 4‐point scale (0–3). Internal consistency in this sample was satisfactory across all three time points (ω = 0.83–0.87).
3.5.1. Coping Self‐Reflection
Coping self‐reflection was measured using a coping self‐reflection scale (Bucknell et al. 2022). This 5‐item scale assessed how participants responded to a recent stressful situation (e.g., “paying attention to what I am doing to manage the situation,” “evaluating how I approached the stressful situation”). Items were rated on a 5‐point scale from 1 (Not at all) to 5 (A lot), with higher scores indicating greater engagement in coping self‐reflection. Internal consistency was excellent across all three time points (ω = 0.91–0.95), and confirmatory factor analysis (CFA) indicated adequate model fit at each time point (Supporting Information S1: Table S1).
3.5.2. Coping‐Related Insight
The research team assessed coping insight using a recently developed coping insight scale (Crane et al. 2024). The original 27‐item coping insight scale was shortened to 14 items, selecting those with the highest factor loadings across five insight sub‐dimensions: (i) Anticipated Efficacy of Resilient Capacities ‐ understanding which strategies and resources are likely to be effective for future challenges (3 items; e.g., “I recognize which coping strategies I can apply to future stressors to help me cope well”); (ii) Current Capacity Repertoire ‐ awareness of one's existing coping skills and resources (4 items; e.g., “I know which coping resources [e.g., friends, time] I access during stressful events”); (iii) Time Course of Reactions ‐ recognition of how emotional and cognitive responses unfold over time (3 items; e.g., “I understand that negative feelings are temporary”); (iv) Stressors as Growth Opportunities ‐ viewing stressors as chances for personal growth (2 items; e.g., “I know that even the most difficult events can contribute to who I am in positive ways”); and (v) Relationships Between Reactions ‐ understanding how thoughts, emotions, and behaviors influence each other during coping (2 items; e.g., “I understand that the way I think can affect the way I feel”). Participants rated how true each statement was during periods of stress on a 5‐point scale (1 = not at all true, 5 = very true). Higher scores indicate stronger coping insight. Internal consistency was satisfactory for all subscales (ω/SBr = 0.77–0.88 across time points). Confirmatory analyses indicated that a model with five correlated first‐order dimensions best fit the data, consistent with prior work (Crane et al. 2024; see Supporting Information S1: Table S1).
3.6. Data Analysis
Raw data were exported from REDCap and imported into SPSS (Version 29) and Mplus (Version 8.11).
3.7. Post‐Hoc Power Analysis
Power analyses were conducted after the sample size had already been established. The sample was constrained by the number of first‐time mothers that we were able to access and retain within the time period for data collection determined by grant funding deadlines (Lakens 2022). Monte Carlo simulations were performed using the Simsem package (Pornprasertmanit et al. 2025) in the R environment (R Core Team 2024). The data‐generating model specified strong to moderate autoregressive paths (β = 0.60–0.30), small cross‐lagged effects (β = 0.20), and strong within‐wave residual covariances (r = 0.60), based on averaged parameter estimates obtained from the observed data. Power to detect individual cross‐lagged paths is moderate (58%–67%), but the power to detect both paths in a chain simultaneously was more demanding and less powered (∼39%–41%). Thus, with N = 147, detecting a full temporal cascade (e.g., Reflection→Insight→Depression) was underpowered in the current sample. This result suggests that non‐significant results cannot be refuted with conventional statistical confidence. To achieve conventional power for the nuanced testing of indirect effects, the sample would need to increase to N = 300 (∼79%‐82%) or N = 400 (∼92%‐93%) in future research.
3.7.1. Multi‐Level Modeling of Change Over Time
Changes in coping self‐reflection, coping‐related insight, and depressive symptoms across the three study waves were analyzed using linear mixed‐effects models in SPSS. Each outcome variable was regressed on Time, which was treated as a three‐level categorical fixed effect, allowing the estimation of within‐person differences across assessment points while accounting for between‐person differences via random intercepts. Models were estimated using restricted maximum likelihood (REML) with Satterthwaite‐adjusted degrees of freedom. Random intercepts were specified for participants to account for individual differences in baseline levels, and residual variances were modeled with a diagonal covariance structure to allow heterogeneity across time. Estimated marginal means were compared to examine change over time.
3.7.2. Multi‐Level Mediation Analysis
We used a 2‐1‐1 multilevel mediation model (Hayes and Rockwood 2020) implemented in SPSS MLmed to examine whether coping insight mediated the association between coping self‐reflection and depressive symptoms. Self‐reflection (X), coping‐related insight (M), and depression (Y) were all measured repeatedly and were decomposed into within‐person (person‐centered) and between‐person (cluster‐mean) components. This allowed us to estimate mediation at both levels. Consistent with the MLmed framework, all regression slopes were treated as fixed, and we specified random intercepts only for the mediator and outcome models. Thus, the a‐path (self‐reflection → coping‐related insight) and b‐path (coping‐related insight → depressive symptoms) were assumed to be constant across individuals, and mediation was estimated separately for within‐person and between‐person processes. Participant identifier was used as the clustering variable.
3.7.3. Cross‐Lagged Panel Models for Causal Pathways
To assess directional and reciprocal relationships between coping self‐reflection, coping insight, and depressive symptoms, we used cross‐lagged panel models (CLPM) across three waves in Mplus. This approach tested: (i) simple lagged models, where earlier self‐reflection predicts later insight, which predicts subsequent depressive symptoms; (ii) reverse causal models, where earlier depression predicts later insight, which influences later self‐reflection; and (iii) reciprocal models, where self‐reflection and depression iteratively influence insight over time. To improve balance in insight change scores, we applied double robustness weighting (Rosenbaum and Rubin 1983), which helped achieve a more representative examination of insight change and increased power to detect nuanced mediation effects (Austin et al. 2018; Shiba and Kawahara 2021).
To compare model fit, we used the Satorra‐Bentler scaled chi‐square difference test (Satorra and Bentler 2010), evaluating: (i) whether lagged models (simple, reverse, and reciprocal) provided better fit than stability models, (ii) whether the reciprocal model fit better than the simple or reverse models. Model fit was assessed using conventional criteria (i.e., Comparative Fit Index (CFI) ≥ 0.95; Tucker‐Lewis Index (TLI) ≥ 0.95; Standardised Root Mean Square Residual (SRMR) ≤ 0.05; Root Mean Square Error of Approximation (RMSEA) ≤ 0.08).
4. Results
4.1. Preliminary Exploration of Data
Descriptive statistics are presented in Tables 1 and 2, and bivariate correlations are reported in Supporting Information S1: Table S2. Depressive symptoms showed strong correlations across time points (Dep T1‐T2: r = 0.706, p < 0.001; Dep T1‐T3: r = 0.462, p < 0.001; Dep T2‐T3: r = 0.454, p < 0.001), indicating temporal stability. Coping self‐reflection was not consistently associated with depressive symptoms, except for a weak negative correlation at Time 3 (r = −0.206, p = 0.011). Each of the five coping insight sub‐dimensions was generally negatively associated with depression across time points, although these links were weak and sometimes non‐significant. Coping self‐reflection showed positive, mostly significant associations with coping insight, with correlations ranging from weak to moderate. Overall, higher reflection was linked to greater coping insight.
TABLE 2.
Descriptive statistics for all variables used in the study.
| Variable | M | SD |
|---|---|---|
| T1 depressive symptoms | 7.56 | 4.86 |
| T2 depressive symptoms | 7.56 | 4.69 |
| T3 depressive symptoms | 8.40 | 5.19 |
| T1 reflection | 3.06 | 0.97 |
| T2 reflection | 2.92 | 0.92 |
| T3 reflection | 2.99 | 1.05 |
| T1 insight into time course of reactions | 3.92 | 0.82 |
| T1 insight into anticipated efficacy of resilient capacities | 3.44 | 0.94 |
| T1 insight into one's current capacity repertoire | 3.52 | 0.88 |
| T1 insight into relationships between emotions, thoughts and behaviors. | 4.08 | 0.69 |
| T1 insight into stressors as opportunities for growth | 3.83 | 0.93 |
| T2 insight into time course of reactions | 3.98 | 0.74 |
| T2 insight into anticipated efficacy of resilient capacities | 3.47 | 0.92 |
| T2 insight into one's current capacity repertoire | 3.54 | 0.80 |
| T2 insight into relationships between emotions, thoughts and behaviors. | 4.10 | 0.67 |
| T2 insight into stressors as opportunities for growth | 3.80 | 0.87 |
| T3 insight into time course of reactions | 4.02 | 0.82 |
| T3 insight into anticipated efficacy of resilient capacities | 3.46 | 0.90 |
| T3 insight into one's current capacity repertoire | 3.55 | 0.82 |
| T3 insight into relationships between emotions, thoughts and behaviors. | 4.12 | 0.78 |
| T3 insight into stressors as opportunities for growth | 3.75 | 0.98 |
4.2. Analysis of Missing Data
Analysis revealed that missing data were minimal, with no significant pattern detected at any time point. There was 2.9% missing data at Time 1 (Little's MCAR test: χ2 (136) = 132.61, p = 0.566), at Time 2 there was 6.85% missing data (Little's MCAR test: χ2 (154) = 133.755, p = 0.879) and at Time 3 there was 8.48% missing data (Little's MCAR test: χ2 (153) = 128.955, p = 0.922). We were able to retain 68% of the sample at Time 3% and 65.33% of the sample completed all time points. For analyses applying a structural equation modeling framework we maximized use of the available data by applying Full Information Maximization Likelihood (FIML) procedure in MPLUS. FIML works with the data that is available to estimate the model parameters, taking into account the missingness (Arbuckle 1996).
4.3. Exploration of Change in Reflection and Coping Insight
We first examined changes in coping self‐reflection, coping insight, and depressive symptoms across the three study waves using multi‐level modeling with a random intercept. Parameter estimates are reported in Table 3. In response to RQ1, results indicated no significant mean differences across time points in coping self‐reflection or the five coping insight sub‐dimensions. However, depressive symptoms significantly increased over time, with higher symptom levels observed at Time 3 (1‐month postpartum) compared to Time 1 (24–32 weeks' gestation) and at T3 (postpartum) compared to T2 (37–39 weeks' gestation).
TABLE 3.
Multi‐level modeling with a random intercept to test mean change across time.
| Effect of time on insight sub‐dimension | ||||||
|---|---|---|---|---|---|---|
| Dependent variable | Est. | SE | t | p | LL | UL |
| Understanding the time course of reactions (reference: T3; 1‐month postpartum) | ||||||
| T1: 24–32 weeks | −0.058 | 0.059 | −0.976 | 0.330 | −0.175 | 0.059 |
| T2: 37–39 weeks | −0.045 | 0.056 | −0.809 | 0.420 | −0.156 | 0.065 |
| Understanding anticipated efficacy of resilient capacities (reference: T3; 1‐month postpartum) | ||||||
| T1: 24–32 weeks | −0.003 | 0.063 | −0.053 | 0.958 | −0.127 | 0.120 |
| T2: 37–39 weeks | 0.003 | 0.063 | 0.044 | 0.965 | −0.121 | 0.127 |
| Understanding one's resilience capacity repertoire (reference: T3; 1‐month postpartum) | ||||||
| T1: 24–32 weeks | 0.117 | 0.073 | 1.615 | 0.108 | −0.026 | 0.261 |
| T2: 37–39 weeks | 0.061 | 0.070 | 0.869 | 0.386 | −0.077 | 0.199 |
| Understanding the relationship between reactions (reference: T3; 1‐month postpartum) | ||||||
| T1: 24–32 weeks | −0.010 | 0.057 | −0.173 | 0.863 | −0.122 | 0.102 |
| T2: 37–39 weeks | −0.026 | 0.056 | −0.466 | 0.642 | −0.136 | 0.084 |
| Understanding stressors as growth opportunities across time (reference: T3; 1‐month postpartum) | ||||||
| T1: 24–32 weeks | 0.117 | 0.073 | 1.615 | 0.108 | −0.026 | 0.261 |
| T2: 37–39 weeks | 0.061 | 0.070 | 0.869 | 0.386 | −0.077 | 0.199 |
| Coping self‐reflection (reference: T3; 1‐month postpartum) | ||||||
| T1: 24–32 weeks | 0.061 | 0.083 | 0.730 | 0.466 | −0.103 | 0.225 |
| T2: 37–39 weeks | −0.071 | 0.081 | −0.876 | 0.382 | −0.230 | 0.088 |
| Depressive symptoms (reference: T3; 1‐month postpartum) | ||||||
| T1: 24–32 weeks | −1.082 | 0.406 | −2.666 | 0.008 | −1.883 | −0.281 |
| T2: 37–39 weeks | −0.822 | 0.404 | −2.032 | 0.044 | −1.620 | −0.024 |
Note: Bolded values indicate statistical significance at p < 0.05.
4.4. Multi‐Level Mediation Analysis of Depressive Symptoms
Five multi‐level models were tested, each with a different coping insight sub‐dimension as the mediator. Results (see Table 4) were consistent across models and supported the Systematic Self‐Reflection Model. Coping self‐reflection was positively associated with all five coping‐related insight dimensions at both within‐ and between‐person levels, while coping‐related insight was negatively linked to depressive symptoms. Consistent with H1, coping‐related insight significantly mediated the relationship between self‐reflection and depressive symptoms across all models: greater self‐reflection predicted higher coping‐related insight, which in turn was associated with lower depressive symptoms.
TABLE 4.
Mediation model results for depression and each of the coping self‐insights as mediators of the relationship between coping self‐reflection and depression.
| Coping self‐reflection → time course of reactions insight→ depressive symptoms | ||||||
|---|---|---|---|---|---|---|
| Within subjects effects | Est. | SE | t | p | LL | UL |
| Reflection (X) →Time course of reactions insight (M) | 0.133 | 0.039 | 3.433 | 0.001 | 0.057 | 0.210 |
| Reflection (X) →Depression (Y) | −0.365 | 0.256 | −1.426 | 0.155 | −0.868 | 0.138 |
| Time course of reactions insight (M)→Depression (Y) | −1.818 | 0.368 | −4.939 | < 0.001 | −2.542 | −1.094 |
| Between subjects effects | Est. | SE | t | p | LL | UL |
|---|---|---|---|---|---|---|
| Reflection (X) → time course of reactions insight (M) | 0.296 | 0.058 | 5.108 | < 0.001 | 0.182 | 0.411 |
| Reflection (X) → depression (Y) | 0.249 | 0.354 | 0.704 | 0.482 | −0.448 | 0.946 |
| Time course of reactions insight (M) → depression (Y) | −2.674 | 0.402 | −6.647 | < 0.001 | −3.467 | −1.881 |
| Indirect effects | Est. | SE | Z | p | MCLL | MCUL |
|---|---|---|---|---|---|---|
| Indirect effect (within) | −0.242 | 0.087 | −2.781 | 0.005 | −0.430 | −0.091 |
| Indirect effect (between) | −0.792 | 0.197 | −4.022 | < 0.001 | −1.225 | −0.439 |
| Test of indirect contextual effect (between—Within) | −0.550 (−1.006 to −0.149) | |||||
| Coping self‐reflection → anticipated efficacy insight → depressive symptoms | ||||||
|---|---|---|---|---|---|---|
| Within subjects effects | Est. | SE | t | p | LL | UL |
| Reflection (X) → anticipated efficacy insight (M) | 0.267 | 0.041 | 6.508 | < 0.001 | 0.187 | 0.348 |
| Reflection (X) → depression (Y) | −0.280 | 0.276 | −1.017 | 0.310 | −0.823 | 0.262 |
| Anticipated efficacy insight (M) → depression (Y) | −1.204 | 0.356 | −3.381 | 0.001 | −1.905 | −0.504 |
| Between subjects effects | Est. | SE | t | p | LL | UL |
|---|---|---|---|---|---|---|
| Reflection (X) → anticipated efficacy insight (M) | 0.580 | 0.059 | 9.878 | < 0.001 | 0.464 | 0.695 |
| Reflection (X) → depression (Y) | 0.654 | 0.423 | 1.548 | 0.123 | −0.179 | 1.487 |
| Anticipated efficacy insight (M) → depression (Y) | −2.071 | 0.418 | −4.960 | < 0.001 | −2.894 | −1.247 |
| Indirect effects | Est. | SE | Z | p | MCLL | MCUL |
|---|---|---|---|---|---|---|
| Indirect effect (within) | −0.322 | 0.108 | −2.973 | 0.003 | −0.550 | −0.127 |
| Indirect effect (between) | −1.200 | 0.272 | −4.415 | < 0.001 | −1.769 | −0.692 |
| Test of indirect contextual effect (between—Within) | −0.878 (−1.495 to −0.322) | |||||
| Coping self‐reflection → current repertoire Insight→ depressive symptoms | ||||||
|---|---|---|---|---|---|---|
| Within subjects effects | Est. | SE | t | p | LL | UL |
| Reflection (X) → current cap. Repertoire insight (M) | 0.235 | 0.035 | 6.685 | < 0.001 | 0.166 | 0.305 |
| Reflection (X) → depression (Y) | −0.212 | 0.277 | −0.767 | 0.444 | −0.756 | 0.332 |
| Current cap. Repertoire insight (M) → depression (Y) | −1.594 | 0.419 | −3.800 | < 0.001 | −2.419 | −0.768 |
| Between subjects effects | Est. | SE | t | p | LL | UL |
|---|---|---|---|---|---|---|
| Reflection (X) → current cap. Repertoire insight (M) | 0.558 | 0.053 | 10.443 | < 0.001 | 0.452 | 0.663 |
| Reflection (X) → depression (Y) | 1.160 | 0.405 | 2.861 | 0.005 | 0.360 | 1.959 |
| Current cap. Repertoire insight (M) → depression (Y) | −3.064 | 0.430 | −7.128 | < 0.001 | −3.911 | −2.216 |
| Indirect effects | Est. | SE | Z | p | MCLL | MCUL |
|---|---|---|---|---|---|---|
| Indirect effect (within) | −0.375 | 0.115 | −3.276 | 0.001 | −0.615 | −0.169 |
| Indirect effect (between) | −1.708 | 0.291 | −5.869 | < 0.001 | −2.321 | −1.163 |
| Test of indirect contextual effect (between—Within) | −1.334 (−1.99 to −0.743) | |||||
| Coping self‐reflection → relationships between reactions Insight→ depressive symptoms | ||||||
|---|---|---|---|---|---|---|
| Within subjects effects | Est. | SE | t | p | LL | UL |
| Reflection (X) →Relationships between reactions insight (M) | 0.184 | 0.037 | 5.033 | < 0.001 | 0.112 | 0.255 |
| Reflection (X) → depression (Y) | −0.377 | 0.268 | −1.406 | 0.161 | −0.905 | 0.151 |
| Relationships between reactions insight (M) →Depression (Y) | −1.253 | 0.402 | −3.113 | 0.002 | −2.044 | −0.461 |
| Between subjects effects | Est. | SE | t | p | LL | UL |
|---|---|---|---|---|---|---|
| Reflection (X) → relationships between reactions insight (M) | 0.307 | 0.050 | 6.214 | < 0.001 | 0.210 | 0.405 |
| Reflection (X) → depression (Y) | 0.008 | 0.386 | 0.021 | 0.983 | −0.754 | 0.770 |
| Relationships between reactions insight (M) →Depression (Y) | −1.800 | 0.501 | −3.594 | < 0.001 | −2.787 | −0.813 |
| Indirect effects | Est. | SE | Z | p | MCLL | MCUL |
|---|---|---|---|---|---|---|
| Indirect effect (within) | −0.230 | 0.088 | −2.610 | 0.009 | −0.419 | −0.077 |
| Indirect effect (between) | −0.553 | 0.180 | −3.082 | 0.002 | −0.945 | −0.230 |
| Test of indirect contextual effect (between—Within) | −0.3230 (−0.7527 to 0.0452) | |||||
| Coping self‐reflection→ stressors as growth opportunities insight → depressive symptoms | ||||||
|---|---|---|---|---|---|---|
| Within subjects effects | Est. | SE | t | p | LL | UL |
| Reflection (X) →Stressors as growth opportunities insight (M) | 0.170 | 0.047 | 3.629 | < 0.001 | 0.078 | 0.262 |
| Reflection (X) → depression (Y) | −0.426 | 0.262 | −1.624 | 0.105 | −0.941 | 0.090 |
| Stressors as growth opportunities insight (M) → depression (Y) | −1.069 | 0.311 | −3.434 | 0.001 | −1.682 | −0.457 |
| Between subjects effects | Est. | SE | t | p | LL | UL |
|---|---|---|---|---|---|---|
| Reflection (X) → stressors as growth opportunities insight (M) | 0.379 | 0.065 | 5.803 | < 0.001 | 0.250 | 0.508 |
| Reflection (X) → depression (Y) | −0.003 | 0.382 | −0.009 | 0.993 | −0.757 | 0.751 |
| Stressors as growth opportunities insight (M) → depression (Y) | −1.429 | 0.381 | −3.756 | < 0.001 | −2.179 | −0.679 |
| Indirect effects | Est. | SE | Z | p | MCLL | MCUL |
|---|---|---|---|---|---|---|
| Indirect effect (within) | −0.182 | 0.074 | −2.446 | 0.014 | −0.344 | −0.057 |
| Indirect effect (between) | −0.542 | 0.174 | −3.121 | 0.002 | −0.924 | −0.233 |
| Test of indirect contextual effect (between—Within) | −0.3602 (−0.769 to −0.018) | |||||
The consideration of contextual effects (Test of Indirect Contextual Effect) in Table 4 reveals a consistent pattern: between‐person differences in reflection amplify the negative indirect effect on depression across multiple insight‐related mediators. Consistent with H1, the indirect relationship between coping self‐reflection and depressive symptoms is significantly stronger at the between‐person level than at the within‐person level. Therefore, stable, trait‐like differences in coping self‐reflection (between‐person effects) more strongly predicted lower depressive symptoms via coping‐related insight than did short‐term, within‐person fluctuations, matching the predicted stronger between‐level effects.
Among the different forms of coping‐related insight, Current Capacity Repertoire exhibited the strongest contextual effect, suggesting that individuals who consistently experience this type of coping‐related insight also experience the most pronounced negative indirect relationship with depressive symptoms. This implies that a stable understanding of one's repertoire of current coping strategies and responses may be especially important for long‐term mental health, whereas short‐term fluctuations in this insight are less predictive of depressive symptoms. In contrast, insight into Relationships between Reactions (e.g., emotions, thoughts, and behaviors) had a non‐significant contextual effect, indicating that between‐person and within‐person indirect effects were of similar magnitude. This pattern reflects a relatively smaller between‐person effect, suggesting that both short‐term changes and stable differences play comparable roles for this insight dimension, unlike other forms where stable differences play a more dominant role.
4.5. Cross‐Lagged Panel Models of Depression to Determine Plausible Causal Chains
To examine causal pathways using cross‐lagged panel models, we first assessed changes in insight levels across Time 1 to Time 2 and Time 2 to Time 3. Given limited and uneven variance in insight change, we applied double robustness weighting to address sampling bias and ensure better representation of coping‐related insight gains and reductions. Coping‐related insight change scores were categorized using a 0.5 SD cutoff, producing nine change groups. Proportional weights (capped at 10×) were applied to balance group representation. The resulting weighted sample showed roughly equal distribution across change categories (ranging from 9.8% to 13.4%), providing adequate balance and improved statistical power to detect subtle cross‐lagged effects (e.g., paths as small as r > 0.2).
Given that proportional weighting simulates a larger sample for certain coping‐related insight change categories, we also present unweighted model results as a sensitivity analysis (Supporting Information S1: Table S3). These models showed minimal change over time in insight autoregressive paths, underscoring the need to balance change rates. As described above we fitted four types of weighted models. These models were tested across all five coping‐related insight dimensions, resulting in 20 total models (see Supporting Information S1: Table S4). Figures 2, 3, 4, 5, 6 display the best‐fitting weighted models for each insight dimension. All alternative models, forward, reversed, and reciprocal, showed significant improvements over the stability model (autoregressive effects only).
FIGURE 2.
Reciprocal cross‐lagged panel model for the time course of reactions insight sub‐domain as a mediator. Effects presented are standardized and significant paths are represented by unbroken lines, whereas non‐significant paths are dashed.
FIGURE 3.
Reciprocal cross‐lagged panel model for the Anticipated Efficacy of Resilience Capacities insight sub‐domain as a mediator. Effects presented are standardized and significant paths are represented by unbroken lines, whereas non‐significant paths are dashed.
FIGURE 4.
Reciprocal cross‐lagged panel model for the Current Capacity Repertoire insight sub‐domain as a mediator. Effects presented are standardized and significant paths are represented by unbroken lines, whereas non‐significant paths are dashed.
FIGURE 5.
Reciprocal cross‐lagged panel model for the relationship between reactions insight sub‐domain as a mediator. Effects presented are standardized and significant paths are represented by unbroken lines, whereas non‐significant paths are dashed.
FIGURE 6.
Reciprocal cross‐lagged panel model for the stressors as growth opportunities insight sub‐domain as a mediator. Effects presented are standardized and significant paths are represented by unbroken lines, whereas non‐significant paths are dashed.
In summary, across the models, reciprocal or reverse models consistently fit better than simple lagged or stability models. Notably, in four cases, the reciprocal models provided a better fit to the data than either the simple or reversed models, as confirmed by chi‐square difference tests (Supporting Information S1: Table S4). For the Relationships Between Reactions model, the reverse‐lagged model fit the data equally well compared with the reciprocal model (SBΔ χ2 = 7.68, p < 0.001). Therefore, we report the reversed lagged model for this insight, while the reciprocal models were reported for all other insight sub‐dimensions. Although reciprocal models most often provided the best fit, they did not produce the reversed indirect effects predicted in H2. Only one coping‐related insight dimension (Stressors as Growth Opportunities) showed the hypothesized indirect effect from reflection to later depressive symptoms via coping‐related insight. In contrast, four of the five insight dimensions showed evidence of a reversed pathway: earlier depressive symptoms predicted lower insight, which in turn predicted lower later coping reflection. Thus, depression‐related inhibition of insight appears to be a more robust temporal mechanism than self‐reflection‐driven reductions in depression. Full results of weighted models are presented below and summarized in Table 5 and fit statistics are in Table S4.
TABLE 5.
Standardised path estimates (β) for the weighted coping‐related insight selected models where model fit was sound.
| Path (reciprocal model) | Estimate (β) | p‐value |
|---|---|---|
| T1 time course of reactions insight →T2 reflection | 0.107 | < 0.001 |
| T2 time course of reactions insight →T3 reflection | 0.141 | < 0.001 |
| T1 reflection →T2 time course of reactions insight | 0.068 | 0.002 |
| T2 reflection →T3 time course of reactions insight | 0.133 | < 0.001 |
| T1 depression →T2 time course of reactions insight | −0.155 | < 0.001 |
| T2 depression →T3 time course of reactions insight | 0.053 | 0.005 a |
| T1 time course of reactions insight →T2 depression | −0.058 | 0.006 |
| T2 time course of reactions insight →T3 depression | −0.056 | 0.042 |
| Indirect: T1 reflection → T2 time course of reactions insight → T3 depression | −0.004 | 0.097 |
| Indirect: T1 depression → T2 time course of reactions insight →T3 reflection | −0.022 | < 0.001 |
| T1 anticipated efficacy of resilient capacities insight → T2 reflection | 0.172 | < 0.001 |
| T2 anticipated efficacy of resilient capacities insight →T3 reflection | 0.239 | < 0.001 |
| T1 reflection → T2 anticipated efficacy of resilient capacities insight | 0.001 | 0.969 |
| T2 reflection → T3 anticipated efficacy of resilient capacities insight | −0.018 | 0.487 |
| T1 anticipated efficacy of resilient capacities insight →T2 depression | −0.078 | < 0.001 |
| T2 anticipated efficacy of resilient capacities insight →T3 depression | −0.222 | < 0.001 |
| T1 depression → T2 anticipated efficacy of resilient capacities insight | −0.092 | < 0.001 |
| T2 depression → T3 anticipated efficacy of resilient capacities insight | 0.115 | < 0.001 |
| Indirect: T1 reflection → T2 anticipated efficacy of resilient cap. → T3 depression | < 0.001 | 0.969 |
| Indirect: T1 depression → T2 anticipated efficacy of resilient cap. → T3 reflection | −0.022 | < 0.001 |
| T1 current capacity repertoire insight → T2 reflection | 0.191 | < 0.001 |
| T2 current capacity repertoire insight → T3 reflection | 0.276 | < 0.001 |
| T1 Reflection→ T2 current capacity repertoire insight | −0.020 | 0.326 |
| T2 Reflection→ T3 current capacity repertoire insight | 0.046 | 0.040 |
| T1 current capacity repertoire insight →T2 depression | −0.108 | < 0.001 |
| T2 current capacity repertoire insight →T3 depression | −0.227 | < 0.001 |
| T1 depression →T2 current capacity repertoire insight | −0.151 | < 0.001 |
| T2 depression →T3 current capacity repertoire insight | 0.129 | < 0.001 |
| Indirect: T1 reflection →T2 current capacity repertoire insight →T3 depression | 0.005 | 0.338 |
| Indirect: T1 depression →T2 current capacity repertoire insight →T3 reflection | −0.042 | < 0.001 |
| T1 relationships between reactions insight →T2 reflection | 0.101 | < 0.001 |
| T2 relationships between reactions insight →T3 reflection | 0.135 | < 0.001 |
| T1 reflection →T2 relationships between reactions insight | — | — |
| T2 reflection →T3 relationships between reactions insight | — | — |
| T1 relationships between reactions insight → T2 depression | — | — |
| T2 relationships between reactions insight → T3 depression | — | — |
| T1 depression →T2 relationships between reactions insight | −0.024 | < 0.001 |
| T2 depression →T3 relationships between reactions insight | 0.004 | 0.267 |
| Indirect: T1 reflection → T2 relationships between reactions insight →T3 depression | — | — |
| Indirect: T1 depression → T2 relationships between reactions insight →T3 reflection | −0.042 | < 0.001 |
| T1 stressors as growth opportunities insight →T2 reflection | 0.075 | 0.004 |
| T2 stressors as growth opportunities insight →T3 reflection | 0.172 | < 0.001 |
| T1 reflection →T2 stressors as growth opportunities insight | 0.110 | < 0.001 |
| T2 reflection →T3 stressors as growth opportunities insight | 0.037 | 0.065 |
| T1 stressors as growth opportunities insight →T2 depression | 0.022 | 0.296 |
| T2 stressors as growth opportunities insight →T3 depression | −0.119 | < 0.001 |
| T1 depression →T2 stressors as growth opportunities insight | −0.039 | 0.080 |
| T2 depression →T3 stressors as growth opportunities insight | 0.003 | 0.887 |
| Indirect: T1 reflection →T2 stressors as growth opportunities insight →T3 depression | −0.013 | 0.006 |
| Indirect: T1 depression →T2 stressors as growth opportunities insight →T3 reflection | −0.007 | 0.093 |
Likely suppression effect; interpret with caution (see main text).
4.5.1. Time Course of Reactions Insight
The reciprocal cross‐lagged model for the Time Course of Reactions insight demonstrated acceptable overall fit (χ 2 (10) = 51.97, p < 0.001; CFI = 0.99; TLI = 0.96; SRMR = 0.022; RMSEA = 0.06). This model was retained as the preferred model (Figure 2). There was clear evidence of a reciprocal association between self‐reflection and the Time Course of Reactions insight. Higher Time Course of Reactions insight at each time point predicted greater self‐reflection at subsequent assessments, and higher self‐reflection likewise predicted increases in this insight over time. Regarding the insight–depression relationship, the pattern was predominantly reciprocal and negative, such that higher insight generally forecast lower subsequent depressive symptoms, and higher depressive symptoms predicted reduced subsequent insight. One statistically positive cross‐lag from T2 depression to T3 insight was identified; however, given the strong autoregressive and concurrent predictors in the model, this pattern is most plausibly attributable to a suppression effect rather than a meaningful reversal in the real‐world relationship. Contrary to H2 (bidirectional indirect effects), the mediation findings were not bidirectional for this insight sub‐dimension. The hypothesized forward indirect effect from coping self‐reflection to later depression via Time Course of Reactions insight was not statistically significant (β = −0.004, p = 0.097), indicating that early differences in coping self‐reflection did not translate into downstream reductions in depressive symptoms through this insight pathway. In contrast, the reverse indirect pathway from depression to later coping self‐reflection was statistically supported. Higher T1 depressive symptoms predicted lower T3 coping self‐reflection through reduced T2 Time Course of Reactions insight (β = −0.022, p < 0.001).
4.5.2. Anticipated Efficacy of Resilient Capacities Insight
The reciprocal model for the Anticipated Efficacy of Resilient Capacities insight sub‐dimension demonstrated good global fit (χ 2 (10) = 36.95, p < 0.001; CFI = 0.99; TLI = 0.98; SRMR = 0.02; RMSEA = 0.05). This model was retained as the preferred specification for examining temporal associations (Figure 3). Results indicated a positive unidirectional relationship between the Anticipated Efficacy of Resilient Capacities insight sub‐dimension and coping self‐reflection. Greater insight at earlier time points predicted higher coping self‐reflection at subsequent waves, whereas coping self‐reflection did not predict later Anticipated Efficacy of Resilient Capacities insight. The associations between Anticipated Efficacy of Resilient Capacities insight and depressive symptoms were largely reciprocal and negative. Higher Anticipated Efficacy of Resilient Capacities insight at T1 and T2 predicted lower depressive symptoms at subsequent time points. Conversely, higher T1 depressive symptoms predicted lower T2 Anticipated Efficacy of Resilient Capacities insight. As in the previous model, a positive association from T2 depression to T3 insight emerged; however, again given the presence of strong autoregressive and concurrent predictors, this pattern is likely driven by suppression rather than reflecting a substantive shift in the underlying relationship. Inconsistent with H2, the indirect effects were not bidirectional. The hypothesized forward indirect effect from coping self‐reflection to later depression, operating through Anticipated Efficacy of Resilient Capacities Coping insight, was not statistically significant (β < 0.001, p = 0.969). Thus, early coping self‐reflection did not translate into downstream reductions in depressive symptoms via this insight pathway. In contrast, the reverse indirect pathway, from early depressive symptoms to later coping self‐reflection through this insight sub‐dimension, was supported. Higher T1 depressive symptoms predicted lower T3 coping self‐reflection via reductions in T2 Anticipated Efficacy of Resilient Capacities insight (β = −0.022, p < 0.001). This suggests that early depressive symptoms may inhibit the emergence of insights regarding the anticipated effectiveness of one's future coping capacities, which in turn constrains later coping self‐reflection.
4.5.3. Current Capacity Repertoire Insight
The reciprocal model for the Current Capacity Repertoire insight sub‐dimension provided the best fit to the data (χ 2 (10) = 88.25, p < 0.001; CFI = 0.98; TLI = 0.94; SRMR = 0.04; RMSEA = 0.08). This model was retained as the preferred model (Figure 4). There was evidence of a reciprocal association between self‐reflection and the Current Capacity Repertoire insight sub‐dimension. Greater insight predicted higher coping self‐reflection at subsequent waves, and coping self‐reflection also predicted later insight. The associations between Current Capacity Repertoire insight and depressive symptoms were reciprocal and negative. Higher Current Capacity Repertoire insight at earlier time points predicted lower depressive symptoms at subsequent waves, consistent with a protective function for awareness of one's existing resilient capacities. Conversely, elevated depressive symptoms at T1 predicted reductions in insight at T2, suggesting that experiencing depressive symptoms may inhibit the ability to recognize one's current coping resources. As in previous models, a positive association from T2 depression to T3 insight emerged, which is likely attributable to suppression given strong autoregressive and concurrent effects rather than reflecting a meaningful shift in the underlying relationship. Inconsistent with H2, the indirect effects were not bidirectional. The hypothesized forward indirect pathway from coping self‐reflection to later depression, operating through Current Capacity Repertoire insight, was not statistically significant (β = 0.006, p = 0.338). Thus, early coping self‐reflection did not predict later depressive symptoms via changes in this insight dimension. In contrast, the reverse indirect effect, from early depressive symptoms to later coping self‐reflection through Current Capacity Repertoire insight, was statistically supported. Higher T1 depressive symptoms predicted lower T3 coping self‐reflection operating via reduced T2 Current Capacity Repertoire insight (β = ‐ 0.042, p < 0.001). This suggests that depressive symptoms impair individuals' ability to recognize their existing coping capacities, which in turn limits later reflective coping.
4.5.4. Relationships Between Reactions Insight
The reverse lagged model for the Relationships Between Reactions insight sub‐dimension was the best model (Figure 5) and an acceptable fit to the data, as indicated by the fit indices (χ2 (14) = 53.98, p < 0.001; CFI = 0.99; TLI = 0.97; SRMR = 0.02; RMSEA = 0.05). The association between insight and reflection occurred in a single direction, whereby higher levels of Relationships Between Reactions insight at the preceding time points were positively related to coping self‐reflection at a subsequent time point. T1 depressive symptoms were negatively associated with the Relationships Between Reactions insight sub‐dimension at T2, but this association did not occur between T2 depressive symptoms to T3 insight and the reverse associations were also not observed. Inconsistent with H2, there was a significant indirect effect of T1 depressive symptoms on T3 coping self‐reflection via the Relationships Between Reactions insight sub‐dimension at T2 (β = −0.023, p < 0.001), but no indirect effect in the reverse direction (Supporting Information S1: Table S4). This suggests that depressive symptoms at an earlier time point indirectly influenced later reflection through its effect on the Relationships Between Reactions insight sub‐dimension.
4.5.5. Stressors as Growth Opportunities Insight
Finally, we tested models where the Stressors as Growth Opportunities insight sub‐dimension served as the mediator. The reciprocal model was the best fit for the data and demonstrated acceptable fit (χ 2 (10) = 54.85, p < 0.001; CFI = 0.99; TLI = 0.95; SRMR = 0.023; RMSEA = 0.062). This model was therefore retained as the preferred model (Figure 6). Partially consistent with H1, greater Stressors as Growth Opportunities insight at earlier time points predicted higher coping self‐reflection at subsequent waves. While T1 coping reflection was positively associated with the T2 Stressors as Growth Opportunities insight sub‐dimension, this same relationship was not observed between T2 coping reflection and T3 insight. The association between the Stressors as Growth Opportunities insight sub‐dimension and depressive symptoms only occurred between T2 insight and T3 depressive symptoms. Inconsistent with H2, the indirect effect between T1 coping reflection and T3 depressive symptoms via T2 Stressors as Growth Opportunities insight sub‐dimension was significant (β = −0.013, p = 0.006) but were not statistically significant in the reverse direction (β = −0.007, p = 0.093).
5. Discussion
5.1. Summary of Findings
In response to RQ1, there was no mean change in insight or reflection across time points with a small, statistically significant increase in depression postnatally (T3) compared to other time points (T1, T2). This stability suggests that coping self‐reflection and insight function as relatively trait‐like metacognitive processes, consistent with prior conceptualisations of private self‐consciousness and self‐focused attention (Trapnell and Campbell 1999). At the same time, first‐time mothers may vary in how much these capacities shift depending on their unique circumstances. Factors such as stress levels, social support, instability, and accessing professional services may promote individual‐level change despite the average pattern demonstrating stability.
Using the multi‐level mediation models, we explored within‐ and between‐person effects. In line with H1, we found that coping‐related insight mediated the relationship between self‐reflection and depressive symptoms at both within‐ and between‐person levels. Higher self‐reflection was positively associated with coping‐related insight, which in turn was linked to lower depressive symptoms, suggesting that self‐reflection is related to mental health primarily via coping‐related insight rather than a direct effect. At the within‐person level, times when individuals reflected more than usual were associated with greater coping‐related insight, which in turn was associated with lower depressive symptoms. However, indirect effects were stronger at the between‐person level. Those who, on average, reflected more reported lower overall depressive symptoms via insight sub‐dimensions. This is consistent with studies demonstrating individuals differ in their dispositional motivation to seek self‐insight thought to drive self‐reflective practice (Heine et al. 2025). Together, this indicates that stable, trait‐like patterns of self‐reflection that support greater average insight may be more protective than short‐term increases in reflection.
Although the multilevel mediation models identified significant between‐person indirect associations, they could not determine the direction of effects. We therefore used cross‐lagged panel models to examine the temporal ordering among variables. Reciprocal models frequently provided the best fit; however, contrary to H2, we found no consistent bidirectional indirect effects among coping self‐reflection, coping‐related insight, and depressive symptoms. At the level of discrete pathways, there was partial evidence of reciprocal associations between coping self‐reflection and coping‐related insight. Bidirectional relationships emerged in three of the five insight sub‐dimensions (Time Course of Reactions, Current Capacity Repertoire, and Stressors as Growth Opportunities) but were not consistent across all time points. For associations with depressive symptoms, three of five insight models (Time Course of Reactions, Current Capacity Repertoire, Anticipated Efficacy of Resilient Capacities) showed reciprocal effects, again with some inconsistency over time. Most notably, a reverse indirect pattern emerged across several models: higher T1 depressive symptoms tended to suppress later coping‐related insight, which in turn reduced later coping self‐reflection. The exception was the Stressors as Growth Opportunities insight model, which showed only a forward indirect effect, suggesting that this form of insight may help translate early coping self‐reflection into later reductions in depressive symptoms.
5.2. Theoretical and Empirical Contributions
5.2.1. Coping Self‐Reflection and Coping‐Related Insight Show Stability
Although we anticipated that metacognitive processes might show greater flexibility during transitional periods, such as the transition to first‐time motherhood, this was not observed in our data. The metacognitive processes measured in this study were relatively stable individual differences. This is consistent with previous work exploring the relationships between metacognitive beliefs and depression. Faissner et al. (2018) showed that metacognitive beliefs are remarkably stable over 3.5 years. This supports the proposal that metacognitive beliefs (i.e., psychological structures, understandings, and processes that are involved in the control, modification, and interpretation of thinking itself, Wells and Cartwright‐Hatton 2004) are relatively stable across time. Having noted this, our findings do not align with prior work showing that instability in life circumstances is associated with a greater motivation for self‐insight (Heine et al. 2025). The timeline for measuring instability may account for the differing findings. In Heine et al. (2025), instability was assessed retrospectively over the previous 12 months, whereas in our study the instability was ongoing at the time of measurement. It may be that the novel stressors accompanying major transitions provoke greater reflection only once the stressor has subsided because the cognitive demands of coping with the present stressor limit the potential for coping self‐reflection. Another possibility is that reflection requires psychological distance; individuals may need time to appraise events meaningfully once there is an outcome from coping attempts. This stability suggests that meaningful increases in reflective tendencies may not occur naturally during pregnancy, or in a timely way, and may instead require targeted intervention. Approaches specifically designed to enhance reflective capacity may therefore be necessary for individuals with lower baseline levels of coping self‐reflection (Crane, Boga, et al. 2019; Falon et al. 2021).
5.2.2. Translation of Coping Self‐Reflection Into Coping‐Related Insight
Further, as predicted by the Systematic Self‐Reflection Model of Resilience Strengthening (Crane, Boga, et al. 2019), it was anticipated that coping self‐reflection would predict greater insight at a later point, potentiating growth across time. Although this pattern appeared in two models, the findings were inconsistent: self‐reflection predicted some forms of insight (e.g., understanding the time course of reactions, perceiving growth opportunities) but not others. This inconsistency suggests that moderating factors may influence the extent to which reflection translates into insight. Potential moderators include ruminative thinking styles, which can divert reflection into unproductive cycles (Bucknell et al. 2022), as well as stressor characteristics, such as its novelty or perceived criticality.
5.2.3. Depressive Symptoms Inhibit Coping‐Related Insight Formation
Across multiple models, early depressive symptoms reduced later insight, which then lowered coping self‐reflection perhaps implicating a vulnerability mechanism. We observed bidirectional associations between depressive symptoms and three insight dimensions: (i) Time Course of Reactions, (ii) Anticipated Efficacy of Resilient Capacities, and (iii) Current Capacity Repertoire. Depression is known to impair cognitive flexibility, reduce attentional control, and narrow interpretive processing (Gotlib and Joormann 2010; Joormann and Gotlib 2010); these effects may limit an individual's capacity to generate insight about stressors or their own coping responses. When insight is diminished, opportunities for coping self‐reflection also decrease, potentially preventing individuals from engaging in the type of reflective processing that supports learning, reappraisal, and emotional regulation. Theoretically, this aligns with models of depression that emphasize disruptions in metacognitive monitoring and control. For example, the Self‐Regulatory Executive Function model (Wells and Matthews 1996) proposes that depression sustains a cycle of rumination that drains the executive resources needed for metacognitive monitoring and reinterpretation. As a result, depressive symptoms may directly hinder the development of coping‐related insight by constraining the system that normally supports reflective reappraisal. It also suggests that insight may function as a mediating mechanism linking low mood to reduced reflective capacity, highlighting that metacognitive processes are not purely trait‐like but can be suppressed by emotional states. This provides an understanding of how depressive symptoms can constrain psychological growth during major life transitions such as pregnancy.
5.2.4. Certain Forms of Coping‐Related Insight Serve a Resilience‐Building Function
Across the longitudinal models, four of the five coping‐related insight domains predicted lower subsequent depressive symptoms: (i) understanding the time course of reactions, (ii) anticipating one's coping efficacy, (iii) recognizing one's current capacity repertoire, and (iv) understanding that stressors may be opportunities for growth. This aligns with previous research demonstrating that greater self‐insight is associated with better mental health outcomes or perceived resilience (Bucknell et al. 2022; Bucknell et al. 2024; Grant et al. 2002; Silvia and Phillips 2011). Just one insight dimension did not have a significant association with depression: Relationships Between Reactions. This is consistent with previous work demonstrating that this sub‐dimension explained the least variance in mental health outcomes (Crane et al. 2024). One possible explanation is that understanding the Relationships Between Reactions reflects declarative knowledge, rather than metacognitive integration based on reflection on one's own coping patterns or self‐understanding. The other four facets all require self‐referential, temporally extended, reflective processing. For example, Time Course of Reactions requires recognizing internal emotional trajectories, anticipating relief, and situating the present self within a changing emotional timeline. Thus, perhaps coping‐related insights that lead to reductions in depressive symptoms must be internalized and self‐referential and generated from one's own experiences and reflection.
Moreover, the four coping‐related insights that were predictive of lower subsequent depressive symptoms also share a common feature in that they are anchored in an emerging understanding of one's control and competence in managing stressors. This constructive, agentic focus likely explains their protective association with depressive symptoms, as such insights may lay the foundation for resilient beliefs like coping self‐efficacy and hope. This interpretation aligns with evidence that reflecting on successes strengthens perceived resilience more than non‐reflective writing (Bucknell, Hoare, et al. 2024), and with recent mixed‐methods findings showing that insights about emotional trajectories, growth opportunities, and personal strengths predict perceived resilience 6 months later. Together, these studies support our observation that the protective effect of coping‐related insights lies in their ability to enhance beliefs about control, competence, and coping capacity.
At the same time, we observed bidirectional associations between depressive symptoms and three insight dimensions. Time Course of Reactions, Anticipated Efficacy of Resilient Capacities, and Current Capacity Repertoire each showed plausible reciprocal relationships, such that while greater insight predicted lower subsequent depression, higher depressive symptoms also undermined these insight capacities. This pattern is consistent with models showing that depression disrupts metacognitive monitoring, impairs the ability to track emotional change over time, constrains future‐oriented coping expectancies, and narrows awareness of available coping resources. Thus, the very insights that protect against depression may be more difficult to access or maintain when depressive symptoms are already present, suggesting a self‐reinforcing cycle in which lowered mood erodes the internal cognitive structures needed for reflective, adaptive coping. Alternatively, these bidirectional associations may reflect not only mutual causal influence but also shared underlying factors (e.g., stress, cognitive load, negative affectivity, or context‐driven constraints on reflection) that simultaneously affect depressive symptoms and insight.
Collectively, these results suggest that coping‐related insight may play a role in protecting against depressive symptoms during the transition to motherhood, while prior depressive symptoms may in turn undermine the development of insight perhaps via effects on impairing cognitive processes or increasing ruminative thought (Bucknell et al. 2022). The stability of reflection and insight across this major life transition suggests these are relatively enduring traits, yet may remain sensitive to individual experiences of stress, adjustment, and emotional well‐being.
5.3. Practical Implications
First, these results highlight the importance of fostering coping‐related insight to support maternal mental health and resilience. Although the dispositional self‐consciousness tradition (Fenigstein et al. 1975) suggests that individuals differ in their stable tendency to attend to aspects of the self, accumulating evidence indicates that self‐reflection is malleable. Experimental studies have successfully evoked self‐reflection by directing attention towards the self through mirrors or video displays (Gendolla et al. 2008; Phillips and Silvia 2005), and intervention research shows that guided self‐reflective journaling can reduce depression and anxiety symptoms (Bucknell, Hoare, et al. 2024; Crane, Boga, et al. 2019). In the present study, the plateauing of coping self‐reflection, reflected in stronger between‐than within‐person effects, suggests that individuals may show trait‐like differences in reflective tendencies, and that sustained, structured intervention may be required to produce meaningful within‐person change. This aligns with reflective‐journaling paradigms that train self‐reflection across extended periods (typically 5 weeks). Taken together, these findings point to the value of structured reflective practices that actively guide new mothers towards constructive, insight‐oriented reflection rather than leaving them to rely on unstructured introspection. Such guidance is essential to ensure that self‐reflection remains adaptive rather than ruminative, a distinction that is particularly important during transitional life phases such as pregnancy and early motherhood (Bucknell et al. 2022; García‐Mieres et al. 2020; Nolen‐Hoeksema 2000).
Second, our findings suggest that depressive symptoms can hinder the development of coping‐related insight, limiting the benefits of spontaneous reflective practice. This indicates that reflection‐based interventions may be most effective for mothers with depressive symptoms when they actively promote critical, constructive insights rather than unstructured introspection. Such approaches could help break the cycle of rumination, enhance coping strategies, and strengthen resilience during the perinatal period.
Third, fostering insights that strengthen a sense of agency and capability appears particularly important for reducing depressive symptoms. Strengths‐based interventions that enhance coping self‐efficacy and reinforce personal agency may help new mothers navigate the challenges of motherhood. By encouraging mothers to recognize and build on their existing coping strengths, these approaches can empower them to reframe stressors as manageable, promote resilience, and reduce feelings of helplessness during this critical life transition.
5.4. Strengths, Limitations, Future Directions for Research
A key strength of this study is its longitudinal, theory‐driven design, which enabled examination of how coping self‐reflection, coping‐related insight, and depressive symptoms unfold across the transition to motherhood. Multiple assessment points allowed us to capture both stability and change, and the combined use of multilevel and cross‐lagged panel models provided insight into within‐person fluctuations and between‐person differences. This is also the first study to apply the Systematic Self‐Reflection Model of Resilience Strengthening to a perinatal context, extending its relevance to maternal mental health.
A primary limitation is that reflection and insight were measured only from mid‐pregnancy to 1 month postpartum, potentially missing earlier or later shifts in these processes. Future research should examine these mechanisms across a broader window, including preconception and several months or years postpartum. Although this study focused on depression, future work should directly test how coping‐related insight contributes to changes in coping strategies, resources, and resilient beliefs, and whether similar patterns emerge for perinatal anxiety. Further research is also needed to clarify which dimensions of coping insight are most protective and how they interact with depression and reflection over time. While some consistency is emerging across studies, the measurement of discrete coping insights is still relatively recent (Crane et al. 2024).
Although rumination is recognized as a key moderator influencing the effects of self‐reflection on coping‐related insight emergence (e.g., Bucknell et al. 2022), it was not included in the present analyses. We did not include rumination as a moderator due to concerns about model complexity and limited statistical power. With a modest sample size (N = 147), the study had reduced power to detect small cross‐lagged effects and was not suited to moderated cross‐lagged models, which subdivide the sample further. Larger samples are needed to examine how maladaptive and adaptive self‐focus jointly shape resilience during the perinatal transition.
Another limitation is the absence of predetermined covariates. Because our aim was to characterize metacognitive coping processes with minimal conditionality, we did not adjust for contextual factors. Nonetheless, unmeasured confounders, such as engagement with psychological services, acute stressors, or social support, may have contributed to observed associations. Future studies should incorporate theoretically selected contextual variables to test the robustness of these processes.
The multilevel models indicated that depressive symptoms showed significant within‐person increases across time, while coping self‐reflection and coping insight dimensions remained largely stable. In this context, standard cross‐lagged panel models can describe temporal associations but may not accurately capture causal effects because they conflate stable between‐person associations with within‐person change, thus their cross‐lagged paths can be biased when trait‐like stability is present (Lucas 2023). These results should be interpreted descriptively rather than as evidence of causal effects. Applying random‐intercept cross‐lagged panel models with a larger sample would better account for the stable individual differences evident in our data.
A further limitation concerns the temporal resolution of our design. The study relied on three assessment points spaced 2 months apart, which cannot capture the fine‐grained, day‐to‐day fluctuations in reflection, insight, and mood during pregnancy. Although meta‐cognitive capacities such as insight tend to be relatively stable across time, daily or momentary assessments could nonetheless clarify how these more trait‐like tendencies interact with short‐term emotional or contextual fluctuations. Such approaches would help disentangle enduring individual differences from state‐level variability and identify whether transient day‐to‐day experiences modulate, amplify, or reduce the broader temporal patterns observed in this study. More intensive longitudinal methods, such as daily diaries, ecological momentary assessment, or measurement‐burst designs, would therefore provide a richer understanding of short‐term processes nested within longer‐term developmental changes.
Although the broader study assessed anxiety, the present analyses focused on depression to maintain conceptual and analytical clarity, given the extensive modeling required to test the Systematic Self‐Reflection Model of Resilience Strengthening. Postpartum depression has received significant empirical attention, with well‐established consequences for maternal functioning and mother–infant outcomes (Biaggi et al. 2016), making it an appropriate starting point for testing the model. Nonetheless, perinatal anxiety is highly prevalent and clinically significant, and future research using this dataset will examine whether similar reflective and insight‐related processes operate in relation to anxiety symptoms.
An additional limitation is that the Coping Self‐Reflection Scale has not yet undergone a dedicated psychometric evaluation. While previous research supports its internal consistency and convergent validity with related constructs (e.g., Bucknell et al. 2022), the absence of comprehensive validation means measurement error cannot be ruled out. Future studies would benefit from formal psychometric assessment of this scale to strengthen confidence in its construct validity and interpretability.
In this study, we focused on new mothers, but additional research is needed to understand resilience in fathers during the transition to parenthood. Fathers are often reluctant to seek support, either avoiding or normalizing mental health concerns, and they frequently have limited support available (Hambidge et al. 2021). Hambidge et al. (2021) found that fathers frequently feel overlooked by healthcare professionals, perceive maternity services as mother‐focused, and struggle with feelings of failure, shame, and isolation when facing mental health challenges. Even fewer studies have examined perinatal mental health outside of heterosexual relationships, leaving significant gaps in understanding and supporting lesbian, bisexual, transgender, and nonbinary parents (Maccio and Pangburn 2011). For example, lesbian and bisexual mothers may face heightened postpartum depression risk due to minority stress, discrimination, and a lack of tailored support. Our study did not assess sexuality or partnering status, which represents an important direction for future work.
6. Conclusion
This study offers new evidence on the role of coping self‐reflection and coping‐related insight in protecting maternal mental health during the transition to motherhood. Coping‐related insight emerged as a key mediator linking self‐reflection and depressive symptoms. While these metacognitive processes are generally stable, they remain sensitive to individual experiences and emotional well‐being. Supporting adaptive coping‐related insight, especially insights that foster agency and capability, may help reduce depressive symptoms and build resilience during this major life transition.
Funding
Funding for research was supported by the Macquarie University COVID Recovery Fellowship Programme (MQRP0001289) and the Australian Research Council Discovery Project Award (DP240100422).
Conflicts of Interest
The authors declare no conflicts of interest.
Supporting information
Supporting Infromation S1
Acknowledgements
Open access publishing facilitated by Macquarie University, as part of the Wiley ‐ Macquarie University agreement via the Council of Australian University Librarians.
Maxwell, Anne‐Marie , Karin Eyal, Scopano Layne, McMahon Cathy, and Crane Monique F.. 2026. “The Relationship Between Depression and Coping Insight Dimensions in the Transition to Motherhood,” Stress and Health: e70144. 10.1002/smi.70144.
Data Availability Statement
The data that support the findings of this study are openly available in Macquarie University Data Repository at https://figshare.com/s/351c0276aabb8b417678.
References
- Ahmed, A. , Bowen A., Feng C. X., and Muhajarine N.. 2019. “Trajectories of Maternal Depressive and Anxiety Symptoms From Pregnancy to Five Years Postpartum and Their Prenatal Predictors.” BMC Pregnancy and Childbirth 19, no. 1: 26. 10.1186/s12884-019-2177-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Aldao, A. , Sheppes G., and Gross J. J.. 2015. “Emotion Regulation Flexibility.” Cognitive Therapy and Research 39, no. 3: 263–278. 10.1007/s10608-014-9662-4. [DOI] [Google Scholar]
- Alves, A. C. , Cecatti J. G., and Souza R. T.. 2021. “Resilience and Stress During Pregnancy: A Comprehensive Multidimensional Approach in Maternal and Perinatal Health.” Scientific World Journal 2021: 1–7. 10.1155/2021/9512854. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Arbuckle, J. L. 1996. “Full Information Estimation in the Presence of Incomplete Data.” In Advanced Structural Equation Modeling. Psychology Press. [Google Scholar]
- Austin, P. C. , Jembere N., and Chiu M.. 2018. “Propensity Score Matching and Complex Surveys.” Statistical Methods in Medical Research 27, no. 4: 1240–1257. 10.1177/0962280216658920. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Bassi, M. , Delle Fave A., Cetin I., et al. 2017. “Psychological Well‐Being and Depression From Pregnancy to Postpartum Among Primiparous and Multiparous Women.” Journal of Reproductive and Infant Psychology 35, no. 2: 183–195. 10.1080/02646838.2017.1290222. [DOI] [PubMed] [Google Scholar]
- Bauer, A. , Knapp M., and Parsonage M.. 2016. “Lifetime Costs of Perinatal Anxiety and Depression.” Journal of Affective Disorders 192: 83–90. 10.1016/j.jad.2015.12.005. [DOI] [PubMed] [Google Scholar]
- Biaggi, A. , Conroy S., Pawlby S., and Pariante C. M.. 2016. “Identifying the Women at Risk of Antenatal Anxiety and Depression: A Systematic Review.” Journal of Affective Disorders 191: 62–77. 10.1016/j.jad.2015.11.014. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Bonanno, G. A. , and Burton C. L.. 2013. “Regulatory Flexibility: An Individual Differences Perspective on Coping and Emotion Regulation.” Perspectives on Psychological Science 8, no. 6: 591–612. 10.1177/1745691613504116. [DOI] [PubMed] [Google Scholar]
- Brockington, I. 2004. “Postpartum Psychiatric Disorders.” Lancet 363, no. 9405: 303–310. 10.1016/S0140-6736(03)15390-1. [DOI] [PubMed] [Google Scholar]
- Bucknell, K. J. , Hoare S., Kangas M., Karin E., and Crane M. F.. 2024. “A mixed‐methods Study Exploring Coping Self‐insights Associated With Resilience.” Behavioral Sciences 14, no. 11: 1018. 10.3390/bs14111018. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Bucknell, K. J. , Kangas M., and Crane M. F.. 2022. “Adaptive Self‐Reflection and Resilience: The Moderating Effects of Rumination on Insight as a Mediator.” Personality and Individual Differences 185: 111234. 10.1016/j.paid.2021.111234. [DOI] [Google Scholar]
- Bucknell, K. J. , Kangas M., Karin E., and Crane M. F.. 2024a. “A Randomized Controlled Trial Comparing the Effects of Self‐Reflective Writing Focused on Successful and Unsuccessful Coping Experiences on Resilience.” Stress and Health 40, no. 2: e3311. 10.1002/smi.3311. [DOI] [PubMed] [Google Scholar]
- Chi, M. T. H. , De Leeuw N., Chiu M., and Lavancher C.. 1994. “Eliciting Self‐Explanations Improves Understanding.” Cognitive Science 18, no. 3: 439–477. 10.1207/s15516709cog1803_3. [DOI] [Google Scholar]
- Ciarrochi, J. , Wilson C. J., Deane F. P., and Rickwood D.. 2003. “Do Difficulties With Emotions Inhibit Help‐Seeking in Adolescence? The Role of Age and Emotional Competence in Predicting Help‐Seeking Intentions.” Counselling Psychology Quarterly 16, no. 2: 103–120. 10.1080/0951507031000152632. [DOI] [Google Scholar]
- Cox, J. L. , Holden J. M., and Sagovsky R.. 1987. “Detection of Postnatal Depression: Development of the 10‐Item Edinburgh Postnatal Depression Scale.” British Journal of Psychiatry 150, no. 6: 782–786. 10.1192/bjp.150.6.782. [DOI] [PubMed] [Google Scholar]
- Crane, M. F. , Boga D., Karin E., et al. 2019. “Strengthening Resilience in Military Officer Cadets: A Group‐Randomized Controlled Trial of Coping and Emotion Regulatory Self‐Reflection Training.” Journal of Consulting and Clinical Psychology 87, no. 2: 125–140. 10.1037/ccp0000356. [DOI] [PubMed] [Google Scholar]
- Crane, M. F. , Hoare S., Kangas M., Gucciardi D. F., and Karin E.. 2024. “A Coping Self‐Insight Scale for Adults: Development and Preliminary Psychometric Properties.” Anxiety, Stress & Coping 37, no. 2: 157–179. 10.1080/10615806.2023.2290099. [DOI] [PubMed] [Google Scholar]
- Crane, M. F. , Searle B. J., Kangas M., and Nwiran Y.. 2019. “How Resilience Is Strengthened by Exposure to Stressors: The Systematic Self‐Reflection Model of Resilience Strengthening.” Anxiety, Stress & Coping 32, no. 1: 1–17. 10.1080/10615806.2018.1506640. [DOI] [PubMed] [Google Scholar]
- Daley, A. , Foster L., Long G., et al. 2015. “The Effectiveness of Exercise for the Prevention and Treatment of Antenatal Depression: Systematic Review With Meta‐Analysis.” BJOG: An International Journal of Obstetrics & Gynaecology 122, no. 1: 57–62. 10.1111/1471-0528.12909. [DOI] [PubMed] [Google Scholar]
- Davis, E. P. , and Narayan A. J.. 2020. “Pregnancy as a Period of Risk, Adaptation, and Resilience for Mothers and Infants.” Development and Psychopathology 32, no. 5: 1625–1639. 10.1017/S0954579420001121. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Davis, R. N. , and Nolen‐Hoeksema S.. 2000. “Cognitive Inflexibility Among Ruminators and Nonruminators.” Cognitive Therapy and Research 24, no. 6: 699–711. 10.1023/a:1005591412406. [DOI] [Google Scholar]
- Dimaggio, G. , Vanheule S., Lysaker P. H., Carcione A., and Nicolò G.. 2009. “Impaired Self‐Reflection in Psychiatric Disorders Among Adults: A Proposal for the Existence of a Network of Semi Independent Functions.” Consciousness and Cognition 18, no. 3: 653–664. 10.1016/j.concog.2009.06.003. [DOI] [PubMed] [Google Scholar]
- Ellis, S. , Carette B., Anseel F., and Lievens F.. 2014. “Systematic Reflection: Implications for Learning From Failures and Successes.” Current Directions in Psychological Science 23, no. 1: 67–72. 10.1177/0963721413504106. [DOI] [Google Scholar]
- Faissner, M. , Kriston L., Moritz S., and Jelinek L.. 2018. “Course and Stability of Cognitive and Metacognitive Beliefs in Depression.” Depression and Anxiety 35, no. 12: 1239–1246. 10.1002/da.22834. [DOI] [PubMed] [Google Scholar]
- Falon, S. L. , Kangas M., and Crane M. F.. 2021. “The Coping Insights Involved in Strengthening Resilience: The Self‐Reflection and Coping Insight Framework.” Anxiety, Stress & Coping 34, no. 6: 734–750. 10.1080/10615806.2021.1910676. [DOI] [PubMed] [Google Scholar]
- Falon, S. L. , Karin E., Boga D., Gucciardi F., Griffin B., and Crane M.. 2021. “A clustered‐randomized Controlled Trial of a Self‐Reflection Resilience‐Strengthening Intervention and Novel Mediators.” Journal of Occupational Health Psychology 26: 1–19. 10.1037/ocp0000268. [DOI] [PubMed] [Google Scholar]
- Farewell, C. V. , Thayer Z., Paulson J., et al. 2022. “Fostering Resilience Among Mothers Early (FRAME): Using Growth Mixture Modeling to Identify Resources That Mitigate Perinatal Depression.” Archives of Women's Mental Health 25, no. 2: 451–461. 10.1007/s00737-022-01211-1. [DOI] [PubMed] [Google Scholar]
- Fenigstein, A. , Scheier M. F., and Buss A. H.. 1975. “Public and Private Self‐Consciousness: Assessment and Theory.” Journal of Consulting and Clinical Psychology 43, no. 4: 522–527. 10.1037/h0076760. [DOI] [Google Scholar]
- Gale, J. , Ooms A., Sharples K., and Marks‐Maran D.. 2016. “The Experiences of Student Nurses on Placements With Practice Nurses: A Pilot Study.” Nurse Education in Practice 16, no. 1: 225–234. 10.1016/j.nepr.2015.08.008. [DOI] [PubMed] [Google Scholar]
- García‐Mieres, H. , De Jesús‐Romero R., Ochoa S., and Feixas G.. 2020. “Beyond the Cognitive Insight Paradox: Self‐Reflectivity Moderates the Relationship Between Depressive Symptoms and General Psychological Distress in Psychosis.” Schizophrenia Research 222: 297–303. 10.1016/j.schres.2020.05.027. [DOI] [PubMed] [Google Scholar]
- Gavin, N. I. , Gaynes B. N., Lohr K. N., Meltzer‐Brody S., Gartlehner G., and Swinson T.. 2005. “Perinatal Depression: A Systematic Review of Prevalence and Incidence.” Obstetrics & Gynecology 106, no. 5 Part 1: 1071–1083. 10.1097/01.AOG.0000183597.31630.db. [DOI] [PubMed] [Google Scholar]
- Gendolla, G. H. E. , Richter M., and Silvia P. J.. 2008. “Self‐Focus and Task Difficulty Effects on Effort‐Related Cardiovascular Reactivity.” Psychophysiology 45, no. 4: 653–662. 10.1111/j.1469-8986.2008.00655.x. [DOI] [PubMed] [Google Scholar]
- Gong, H. , Ni C., Shen X., Wu T., and Jiang C.. 2015. “Yoga for Prenatal Depression: A Systematic Review and meta‐analysis.” BMC Psychiatry 15, no. 1: 14. 10.1186/s12888-015-0393-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Gotlib, I. H. , and Joormann J.. 2010. “Cognition and Depression: Current Status and Future Directions.” Annual Review of Clinical Psychology 6, no. 1: 285–312. 10.1146/annurev.clinpsy.121208.131305. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Grant, A. M. , Curtayne L., and Burton G.. 2009. “Executive Coaching Enhances Goal Attainment, Resilience and Workplace Well‐Being: A Randomised Controlled Study.” Journal of Positive Psychology 4, no. 5: 396–407. 10.1080/17439760902992456. [DOI] [Google Scholar]
- Grant, A. M. , Franklin J., and Langford P.. 2002. “The self‐reflection and Insight Scale: A New Measure of Private Self‐Consciousness.” Social Behavior and Personality: An International Journal 30, no. 8: 821–835. 10.2224/sbp.2002.30.8.821. [DOI] [Google Scholar]
- Hajure, M. , Alemu S. S., Abdu Z., et al. 2024. “Resilience and Mental Health Among Perinatal Women: A Systematic Review.” Frontiers in Psychiatry 15: 1373083. 10.3389/fpsyt.2024.1373083. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Hambidge, S. , Cowell A., Arden‐Close E., and Mayers A.. 2021. ““What Kind of Man Gets Depressed After Having a Baby?” Fathers’ Experiences of Mental Health During the Perinatal Period.” BMC Pregnancy and Childbirth 21, no. 1: 463. 10.1186/s12884-021-03947-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Hannon, S. E. , Daly D., and Higgins A.. 2022. “Resilience in the Perinatal Period and Early Motherhood: A Principle‐Based Concept Analysis.” International Journal of Environmental Research and Public Health 19, no. 8: 4754. 10.3390/ijerph19084754. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Hayes, A. F. , and Rockwood N. J.. 2020. “Conditional Process Analysis: Concepts, Computation, and Advances in the Modeling of the Contingencies of Mechanisms.” American Behavioral Scientist 64, no. 1: 19–54. 10.1177/0002764219859633. [DOI] [Google Scholar]
- He, W. , and Gan J.. 2025. “The Relationship Between Self‐Reflection and Mental Health: A Meta‐Analysis Review.” Current Psychology 44, no. 5: 3899–3913. 10.1007/s12144-025-07415-9. [DOI] [Google Scholar]
- Heine, C. , Schmukle S. C., and Dufner M.. 2025. “The Quest for Genuine self‐knowledge: An Investigation Into Individual Differences in the Self‐Insight Motive.” European Journal of Personality 39, no. 4: 455–477. 10.1177/08902070241272184. [DOI] [Google Scholar]
- Highet, N. and Centre of Perinatal Excellence . 2023. Perinatal Mental Health Practice Guideline, 1–145. Centre of Perinatal Excellence. [Google Scholar]
- Howard, L. M. , and Khalifeh H.. 2020. “Perinatal Mental Health: A Review of Progress and Challenges.” World Psychiatry 19, no. 3: 313–327. 10.1002/wps.20769. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Hu, Y. , Wang Y., Wen S., et al. 2019. “Association Between Social and Family Support and Antenatal Depression: A Hospital‐Based Study in Chengdu, China.” BMC Pregnancy and Childbirth 19, no. 1: 420. 10.1186/s12884-019-2510-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Joormann, J. , and Gotlib I. H.. 2010. “Emotion Regulation in Depression: Relation to Cognitive Inhibition.” Cognition & Emotion 24, no. 2: 281–298. 10.1080/02699930903407948. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Kalisch, R. , Baker D. G., Basten U., et al. 2017. “The Resilience Framework as a Strategy to Combat Stress‐Related Disorders.” Nature Human Behaviour 1, no. 11: 784–790. 10.1038/s41562-017-0200-8. [DOI] [PubMed] [Google Scholar]
- Kato, T. 2012. “Examination of the Coping Flexibility Hypothesis Using the Coping Flexibility Scale‐Revised.” Journal of Counseling Psychology 59, no. 2: 262–273. 10.1037/a0027770. [DOI] [PubMed] [Google Scholar]
- Lakens, D. 2022. “Sample Size Justification.” Collabra: Psychology 8, no. 1: 33267. 10.1525/collabra.33267. [DOI] [Google Scholar]
- Levis, B. , Negeri Z., Sun Y., Benedetti A., and Thombs B. D.. 2020. “Accuracy of the Edinburgh Postnatal Depression Scale (EPDS) for Screening to Detect Major Depression Among Pregnant and Postpartum Women: Systematic Review and Meta‐Analysis of Individual Participant Data.” BMJ m4022: m4022. 10.1136/bmj.m4022. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Lucas, R. E. 2023. “Why the Cross‐Lagged Panel Model Is Almost Never the Right Choice.” Advances in Methods and Practices in Psychological Science 6, no. 1: 1–22. 10.1177/25152459231158378. [DOI] [Google Scholar]
- Maccio, E. M. , and Pangburn J. A.. 2011. “The Case for Investigating Postpartum Depression in Lesbians and Bisexual Women.” Women's Health Issues 21, no. 3: 187–190. 10.1016/j.whi.2011.02.007. [DOI] [PubMed] [Google Scholar]
- McAdams, D. P. 2001. “The Psychology of Life Stories.” Review of General Psychology 5, no. 2: 100–122. 10.1037/1089-2680.5.2.100. [DOI] [Google Scholar]
- McMahon, C. 2023. Becoming a Parent: Contemporary Contexts and Challenges During the Transition to Parenthood. Cambridge University Press. [Google Scholar]
- Munk‐Olsen, T. , Laursen T. M., Pedersen C. B., Mors O., and Mortensen P. B.. 2006. “New Parents and Mental Disorders: A Population‐Based Register Study.” JAMA 296, no. 21: 2582. 10.1001/jama.296.21.2582. [DOI] [PubMed] [Google Scholar]
- Nolen‐Hoeksema, S. 2000. “The Role of Rumination in Depressive Disorders and Mixed Anxiety/Depressive Symptoms.” Journal of Abnormal Psychology 109, no. 3: 504–511. 10.1037/0021-843x.109.3.504. [DOI] [PubMed] [Google Scholar]
- Park, C. L. 2010. “Making Sense of the Meaning Literature: An Integrative Review of Meaning Making and Its Effects on Adjustment to Stressful Life Events.” Psychological Bulletin 136, no. 2: 257–301. 10.1037/a0018301. [DOI] [PubMed] [Google Scholar]
- Phillips, A. G. , and Silvia P. J.. 2005. “Self‐Awareness and the Emotional Consequences of Self‐Discrepancies.” Personality and Social Psychology Bulletin 31, no. 5: 703–713. 10.1177/0146167204271559. [DOI] [PubMed] [Google Scholar]
- Pornprasertmanit, S. , Miller P., Schoemann A. M., and Jorgensen T. D.. 2025. “Simsem: Simulated Structural Equation Modeling.” [Computer software] . https://CRAN.R‐project.org/package=simsem.
- PwC Consulting Australia 2019. The Cost of Perinatal Depression and Anxiety in Australia, 1–26. Price Waterhouse Coopers Consulting: (Australia). [Google Scholar]
- R Core Team 2024. R: A Language and Environment for Statistical Computing.”Version 4.4.1. [Computer software], R Foundation for Statistical Computing. https://www.R‐project.org/. [Google Scholar]
- Richardson, G. E. 2002. “The Metatheory of Resilience and Resiliency.” Journal of Clinical Psychology 58, no. 3: 307–321. 10.1002/jclp.10020. [DOI] [PubMed] [Google Scholar]
- Rosenbaum, P. R. , and Rubin D. B.. 1983. “The Central Role of the Propensity Score in Observational Studies for Causal Effects.” Biometrika 70, no. 1: 41–55. 10.1093/biomet/70.1.41. [DOI] [Google Scholar]
- Satorra, A. , and Bentler P. M.. 2010. “Ensuring Positiveness of the Scaled Difference Chi‐Square Test Statistic.” Psychometrika 75, no. 2: 243–248. 10.1007/s11336-009-9135-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Shiba, K. , and Kawahara T.. 2021. “Using Propensity Scores for Causal Inference: Pitfalls and Tips.” Journal of Epidemiology 31, no. 8: 457–463. 10.2188/jea.JE20210145. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Shields, G. S. , Sazma M. A., and Yonelinas A. P.. 2016. “The Effects of Acute Stress on Core Executive Functions: A Meta‐Analysis and Comparison With Cortisol.” Neuroscience & Biobehavioral Reviews 68: 651–668. 10.1016/j.neubiorev.2016.06.038. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Silvia, P. J. , and Phillips A. G.. 2011. “Evaluating Self‐Reflection and Insight as Self‐Conscious Traits.” Personality and Individual Differences 50, no. 2: 234–237. 10.1016/j.paid.2010.09.035. [DOI] [Google Scholar]
- Snyder, H. R. 2013. “Major Depressive Disorder Is Associated With Broad Impairments on Neuropsychological Measures of Executive Function: A Meta‐Analysis and Review.” Psychological Bulletin 139, no. 1: 81–132. 10.1037/a0028727. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Stein, A. , Pearson R. M., Goodman S. H., et al. 2014. “Effects of Perinatal Mental Disorders on the Fetus and Child.” Lancet 384, no. 9956: 1800–1819. 10.1016/S0140-6736(14)61277-0. [DOI] [PubMed] [Google Scholar]
- Takano, K. , and Tanno Y.. 2009. “Self‐Rumination, Self‐Reflection, and Depression: Self‐Rumination Counteracts the Adaptive Effect of Self‐Reflection.” Behaviour Research and Therapy 47, no. 3: 260–264. 10.1016/j.brat.2008.12.008. [DOI] [PubMed] [Google Scholar]
- Thompson, S. F. , Shimomaeda L., Calhoun R., Moini N., Smith M. R., and Lengua L. J.. 2022. “Maternal Mental Health and Child Adjustment Problems in Response to the COVID‐19 Pandemic in Families Experiencing Economic Disadvantage.” Research on Child and Adolescent Psychopathology 50, no. 6: 695–708. 10.1007/s10802-021-00888-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Trapnell, P. D. , and Campbell J. D.. 1999. “Private Self‐Consciousness and the Five‐Factor Model of Personality: Distinguishing Rumination From Reflection.” Journal of Personality and Social Psychology 76, no. 2: 284–304. 10.1037/0022-3514.76.2.284. [DOI] [PubMed] [Google Scholar]
- Treynor, W. , Gonzalez R., and Nolen‐Hoeksema S.. 2003. “Rumination Reconsidered: A Psychometric Analysis.” Cognitive Therapy and Research. [Google Scholar]
- Wang, Z. , Liu J., Shuai H., et al. 2021. “Mapping Global Prevalence of Depression Among Postpartum Women.” Translational Psychiatry 11, no. 1: 543. 10.1038/s41398-021-01663-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Wells, A. , and Cartwright‐Hatton S.. 2004. “A Short Form of the Metacognitions Questionnaire: Properties of the MCQ‐30.” Behaviour Research and Therapy 42, no. 4: 385–396. 10.1016/S0005-7967(03)00147-5. [DOI] [PubMed] [Google Scholar]
- Wells, A. , and Matthews G.. 1996. “Modelling Cognition in Emotional Disorder: The S‐REF Model.” Behaviour Research and Therapy 34, no. 11–12: 881–888. 10.1016/S0005-7967(96)00050-2. [DOI] [PubMed] [Google Scholar]
- Yehuda, R. , and Meaney M. J.. 2018. “Relevance of Psychological Symptoms in Pregnancy to Intergenerational Effects of Preconception Trauma.” Biological Psychiatry 83, no. 2: 94–96. 10.1016/j.biopsych.2017.10.027. [DOI] [PubMed] [Google Scholar]
- Young, C. , and Ayers S.. 2021. “Risk and Resilience in Pregnancy and Birth.” In Multisystemic Resilience, edited by Ungar M.. 1st ed., 57–77. Oxford University PressNew. 10.1093/oso/9780190095888.003.0004. [DOI] [Google Scholar]
- Zhang, J. , Wang P., Fan W., and Lin C.. 2024. “Comparing the Prevalence and Influencing Factors of Postpartum Depression in Primiparous and Multiparous Women in China.” Frontiers in Psychiatry 15: 1479427. 10.3389/fpsyt.2024.1479427. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Citations
- Pornprasertmanit, S. , Miller P., Schoemann A. M., and Jorgensen T. D.. 2025. “Simsem: Simulated Structural Equation Modeling.” [Computer software] . https://CRAN.R‐project.org/package=simsem.
Supplementary Materials
Supporting Infromation S1
Data Availability Statement
The data that support the findings of this study are openly available in Macquarie University Data Repository at https://figshare.com/s/351c0276aabb8b417678.