The general psychopathology factor (p factor) reflects a common liability underlying a broad range of emotional and behavioural problems in childhood. Although adverse childhood experiences (ACEs) and temperament are established vulnerability dimensions, the mechanisms through which contextual and dispositional influences converge on general psychopathology remain unclear. Emotion regulation may represent a key transdiagnostic process linking early vulnerability to broad psychopathological risk.
In a community sample of 241 children aged 6–12 years (57.7% girls), a path model examined associations among cumulative ACEs, temperament dimensions (negative affectivity, effortful control, behavioural inhibition, and surgency), emotion regulation processes (emotional lability and adaptive regulation), and the p factor. The model was estimated using WLSMV and adjusted for age and sex.
The model explained 45% of the variance in the p factor. Emotion dysregulation was positively associated with general psychopathology (β = 0.510, p < .001), whereas adaptive regulation was not significant. ACEs (β = 0.201, p < .001), behavioural inhibition (β = 0.152, p = .003), and surgency (β = −.100, p = .023) showed direct associations with the p factor. Significant indirect effects through emotional lability were observed for adversity, negative affectivity, effortful control, behavioural inhibition, and surgency.
Emotion dysregulation emerged as a key intermediary process linking adversity and temperament with broad psychopathology in middle childhood. These findings highlight emotion dysregulation as a clinically relevant transdiagnostic target that may help identify children vulnerable to broad psychopathology and guide early preventive interventions.
The general psychopathology factor (p factor) reflects a common latent dimension underlying diverse forms of psychopathology, helping to explain why individuals with one disorder are at increased risk for developing others and capturing broad features such as comorbidity, symptom severity, and persistence over time (Caspi et al., 2014, 2018; Kotov et al., 2017). This common dimension emerges from dimensional models of psychopathology as a response to the limitations of categorical diagnostic systems.
Dimensional approaches show that psychological disorders, despite their heterogeneity, share substantial symptom overlap, high comorbidity, and common vulnerability factors—biological, contextual, and psychological—arising from the dynamic interplay between dispositional characteristics and environmental influences (Caspi et al., 2014; Caspi & Moffitt, 2018; Lynch et al., 2021; Sun et al., 2025).
For instance, Barlow’s (2004) triple vulnerability model illustrates how genetically influenced traits, including temperament, interact with environmental stressors and regulatory processes to confer risk for emotional disorders. Building on this dimensional view, the Hierarchical Taxonomy of Psychopathology (HiTOP) offers an empirically derived hierarchical organization of psychopathology, positioning the p factor at its highest level as an index of general liability across symptom domains (Kotov et al., 2017; Ringwald et al., 2023). In parallel, the Research Domain Criteria (RDoC) framework advances an integrative, developmental approach by linking multiple levels of analysis—ranging from genetic and biological to psychological, behavioural, and environmental processes—to the study of general psychopathology (Michelini et al., 2021).
Although the p factor shows moderate stability, with higher levels predicting increased vulnerability to psychopathology across the lifespan (Caspi et al., 2018; Choate et al., 2023), emerging evidence suggests that it is not a fixed predisposition, particularly during childhood. Accordingly, recent research has focused on identifying the factors associated with general psychopathology, extending earlier work on internalizing and externalizing problems (Brislin et al., 2022; Deutz et al., 2020). In their systematic review, Lynch et al. (2021) identified several biological, contextual, and psychological factors, including temperament traits, as key correlates of the p factor.
Adverse childhood experiences and p factorAdverse childhood experiences (ACEs) are traumatic or stressful events occurring before age 18 and are strongly linked to later physical and psychological difficulties. Extensive reviews indicate that a greater exposure to adversity is associated with increased severity and chronicity of psychopathology across the lifespan, with the timing of these experiences playing a critical role in shaping long-term outcomes (Hawes & Allen, 2023; Hinojosa & Hinojosa, 2024; Kirkbride et al., 2024; Petruccelli et al., 2019; Uher et al., 2023). ACEs are also highly prevalent and disproportionately affect children from socioeconomically disadvantaged, single-parent, and racial or ethnic minority backgrounds, where cumulative exposure to adversity amplifies developmental risk (Adjei et al., 2023; Deighton et al., 2019; Kirkbride et al., 2024; Lanier et al., 2018; O’Hare et al., 2023). Although the relationship between ACEs and psychopathology is well established, its association with the general psychopathology factor (p factor) remains relatively underexplored (Hawes & Allen, 2023).
Lynch et al. (2021) reported that stressful life events predict general psychopathology and externalizing symptoms, whereas exposure to violence shows more specific associations with externalizing problems and less consistent links with either the p factor or internalizing symptoms. Similarly, parental psychopathology has been associated with elevated risk across internalizing, externalizing, and general psychopathology domains, while other forms of adversity show weaker or more variable associations (Liu et al., 2017; McKelvey et al., 2020; Ronis et al., 2023).
Moreover, although early adversity has been consistently associated with elevated developmental risk, children exposed to similar adverse experiences do not uniformly exhibit comparable outcomes, indicating substantial heterogeneity in how different dimensions of adversity, such as threat and deprivation, relate to behavioural and cognitive functioning (Sayler et al., 2025). This heterogeneity suggests that individual differences may shape children’s developmental responses to adversity.
Progress in this area has been limited by substantial methodological heterogeneity, including variability of participant characteristics, heterogeneity in the definition and measurement of ACEs, and differences in the assessment of mental health outcomes (Baldwin et al., 2023; McLaughlin et al., 2023; Petruccelli et al., 2019; Sisitsky et al., 2023; Tzouvara et al., 2023). These limitations constrain broader understanding of how ACEs relate to general psychopathology.
Beyond these challenges, the variability in associations between ACEs and the p factor points to the importance of identifying mechanisms that explain differential susceptibility to adversity and the role of other processes shaping these relationships (Liu et al., 2017). Understanding why some children develop elevated general psychopathology following adverse experiences, whereas others show relative resilience, is critical for advancing transdiagnostic models and informing the development of robust, evidence-based preventive and intervention strategies.
Temperament and p factorTemperament has increasingly been conceptualized as a biologically based vulnerability factor involved in stress responsivity and emotional adjustment. According to the Regulative Theory of Temperament proposed by Strelau (1998), temperament traits reflect relatively stable individual differences in emotional and behavioural reactivity that become particularly relevant under stressful or highly stimulating conditions (Strelau, 1998). In this framework, traits related to emotional reactivity and perseveration may increase vulnerability to maladaptive emotional responses following adverse or traumatic experiences, whereas traits such as endurance or briskness may operate as protective factors facilitating adaptation under stress (Strelau & Zawadzki, 2005). Research on trauma and posttraumatic stress symptoms has consistently shown that exposure to adversity alone does not fully explain psychopathological outcomes, as individual differences in temperament significantly modulate emotional responding and stress-related vulnerability. In particular, heightened emotional reactivity — characterized by increased emotional sensitivity and reduced emotional endurance — has been associated with greater emotional dysregulation and vulnerability to psychopathological symptoms across both clinical and non-clinical populations. These findings support developmental and transdiagnostic perspectives suggesting that temperament may shape how children perceive, respond to, and regulate emotional experiences following exposure to adversity (Rzeszutek & Oniszczenko, 2022; Strelau & Zawadzki, 2005).
Consistent with these perspectives, temperament has been associated with broad dimensions of psychopathology, including the p factor, with several temperament traits prospectively predicting general psychopathological liability across development (Leaberry et al., 2019; Morales et al., 2021; Ormel et al., 2024). From a developmental perspective, temperament is increasingly conceptualized as one of the earliest building blocks of psychopathology. Within the Research Domain Criteria (RDoC) framework, temperament—particularly individual differences in emotional reactivity and regulatory control at the dispositional level—-has been proposed as a set of early-emerging, transdiagnostic vulnerabilities that interact with environmental contexts (Ostlund et al., 2021).
Some researchers have raised concerns about conceptual overlap between certain temperament traits—particularly negative affectivity—and psychopathological symptoms (e.g. Brandes et al., 2019). However, empirical evidence supports the distinction between these constructs and underscores temperament as a core source of individual differences relevant to psychopathology (Barlow & Kennedy, 2016; Hankin et al., 2017; Lynch et al., 2021; Rothbart & Bates, 2006).
Although temperament traits may differentially predispose individuals to specific forms of psychopathology (Nigg, 2006; Rothbart & Bates, 2006), findings based on isolated traits remain inconsistent. Among the temperamental dimensions most frequently examined in relation to psychopathology and the p factor are negative affectivity or neuroticism, effortful control, behavioural inhibition, and surgency (Barlow & Kennedy, 2016; Brandes & Tackett, 2019; Hankin et al., 2017; Lynch et al., 2021). Across studies, elevated negative affectivity and low effortful control have been most consistently associated with higher levels of general psychopathology, including both internalizing and externalizing symptoms, as well as with the p factor, although effect sizes vary considerably (e.g. Shields et al., 2019; Wang et al., 2020; Winebrake et al., 2024). Evidence for other temperament dimensions, such as positive affectivity or surgency, is less consistent. For example, Morales et al. (2021) identified high motor activity—a specific facet of surgency—as a predictor of the p factor, whereas other components of positive affectivity, including low positive affect, appear more closely related to internalizing symptoms than to general psychopathology (Lynch et al., 2021).
Together, this body of evidence suggests that while temperament constitutes a meaningful dispositional foundation for general psychopathological liability, associations between specific traits and the p factor are often modest and heterogeneous. This pattern indicates that temperament may exert its influence on general psychopathology primarily through intermediate emotional processes rather than through direct effects alone.
The role of emotion regulationThe p factor has been linked to both exposure to adversity and temperamental traits, but no single risk factor is sufficient to explain the development of psychopathology, underscoring the need for multifactorial and transdiagnostic models.
Emotion regulation is widely implicated in a range of psychological difficulties and is increasingly conceptualized as a transdiagnostic process that cuts across diagnostic categories and levels of analysis (Aldao et al., 2010; Aldao, 2016; Clamor et al., 2026; Cludius et al., 2020; Fernandez et al., 2016; Liu et al., 2026). However, its specific role in relation to the p factor—whether as a core dispositional vulnerability, a proximal mediating mechanism, or both—remains incompletely understood.
Associations between emotion regulation and psychopathology vary across strategies and symptom domains, with stronger links typically observed for internalizing symptoms. Nevertheless, emotion regulation is often proposed as a key developmental mechanism through which adversity and temperament shape vulnerability to psychopathology.
Empirical studies indicate that children exposed to adverse childhood experiences show elevated psychopathology when emotion regulation capacities are compromised (Cole & Diaz, 2024; Messman-Moore & Bhuptani, 2017; Miu et al., 2022; Zhang et al., 2024), possibly reflecting disruptions in the development of adaptive strategies and increased reliance on maladaptive responses such as rumination or suppression, despite notable methodological limitations in this literature (Hales et al., 2023).
In addition, emotion regulation is also closely linked to temperament. Profiles marked by high negative affectivity and low effortful control are associated with poorer regulatory capacities in both children (Liu et al., 2024; Van Beveren et al., 2016; Verstraeten et al., 2009; Zhang et al., 2023; Zuzama et al., 2020) and adults (Singh & Mishra, 2023; Tortella-Feliu et al., 2012). Longitudinal evidence further indicates that early temperamental risk may be translated into broad psychopathology via regulatory pathways (Deutz et al., 2020).
However, the association between negative affectivity and psychopathology is relatively well established, the intermediary mechanisms through which this dispositional vulnerability translates into transdiagnostic risk remain less clearly understood. Within the hierarchical framework articulated by Brandes and Tackett (2019), negative affectivity is robustly linked to multiple levels of the psychopathology hierarchy, including the general factor. The authors emphasize that considerably less is known about how dispositional traits align with intermediate and lower-order levels of psychopathology. This suggests that observed associations may reflect not only shared dispositional content but also underlying regulatory processes that mediate the expression of vulnerability. Accordingly, identifying intermediary mechanisms—such as emotional regulation—may be essential for integrating hierarchical models of personality and psychopathology and for advancing more mechanistically informative accounts of general psychopathological liability.
Despite growing interest in emotion regulation as a transdiagnostic mechanism, few studies have directly examined its association with the p factor during childhood (e.g., Cai et al., 2024; Weissman et al., 2019). Clarifying the role of modifiable regulatory processes may therefore help explain differential vulnerability to adversity and temperament and inform the development of preventive and intervention strategies targeting general psychopathology.
The current studyThe present study aims to examine how adverse childhood experiences (ACEs), temperament, and emotion regulation jointly contribute to the general psychopathology factor (p factor) in children aged 6–12 years. Specifically, we test a path model integrating adversity, temperament, emotion regulation, and the p factor within a single analytic framework.
This study addresses this issue by examining contextual, dispositional, and regulatory correlates of broad psychopathological liability within a single developmental model. Although adversity, temperament, and emotion regulation have each been independently associated with child psychopathology, fewer studies have considered how these domains are jointly related to the general psychopathology factor during middle childhood. This developmental period is particularly relevant, as individual differences in emotional reactivity and regulatory capacities are still consolidating and may shape children’s broader vulnerability to psychological difficulties. A further contribution of the present study is the distinction between emotional lability and adaptive emotion regulation, which allows us to examine whether p factor is more closely associated with dysregulated affective reactivity than with reported adaptive regulatory capacities. By integrating cumulative adversity, temperament dimensions, and emotion regulation processes in relation to the p factor, this study aims to clarify how contextual and dispositional vulnerability are associated with broad psychopathology in childhood.
In this model, we evaluate the direct associations between exposure to ACEs, emotion regulation (emotional lability and adaptive regulation), and the p factor. Based on prior research, we hypothesise that higher levels of adversity are associated with higher p factor scores and with greater emotional lability and poorer adaptive regulation. Given evidence of heterogeneity in outcomes following adversity, we further expect that the association between ACEs and the p factor is partly accounted for by emotion regulation processes.
In parallel, we examine the associations between four temperamental traits (negative affectivity, effortful control, behavioural inhibition, and surgency), emotion regulation, and the general psychopathology factor (p factor). We hypothesise that higher negative affectivity and behavioural inhibition, and lower effortful control and surgency, are associated with higher p factor scores, greater emotional lability, and lower adaptive emotion regulation. Furthermore, we expect that the associations between temperamental traits and the p factor are largely indirect once emotion regulation processes are considered, consistent with a potential intermediary role of emotion regulation in the statistical model.
MethodParticipantsThe sample consisted of 241 children aged between 6 and 12 years (M = 8.13, SD = 1.95), 57.7% of whom were girls. Data were collected from parents, the majority of whom were mothers (91%), with a mean age of 42.2 years (range = 27–55 years). Participants were recruited from primary schools in Spain and represented a community-based sample of families from the general population, with no prior selection based on clinical characteristics. The sample largely reflected middle-socioeconomic-status families within a relatively homogeneous sociocultural context, as shown by the sociodemographic characteristics reported in the Results section and Supplementary Table S1. Inclusion criteria required that children fell within the target age range and that a primary caregiver had access to a mobile device and was willing to complete online questionnaires in either Catalan or Spanish. The study was approved by the relevant institutional ethics committee, and informed consent was obtained from all participating families prior to data collection.
InstrumentsP factorChildren’s psychopathological symptomatology was assessed using the Child Behavior Checklist (CBCL), developed by Achenbach and Rescorla (2001). It consists of 113 items, scored on a 3-point scale (0 = never, 1 = sometimes, 2 = often), and evaluates multiple dimensions of psychopathology, including internalizing problems, externalizing problems, and thought problems. This study employed the Spanish version of the CBCL, available at https://www.ued.uab.cat/.
Broad psychopathological liability (p factor) was operationalized using the CBCL Total Problems Score as an observed dimensional proxy aligned with the literature on general psychopathology. Although this score does not constitute a latent estimate of the p factor, it captures broad cross-domain symptom burden across internalizing, externalizing, attentional, behavioural, and emotional problems, and may therefore serve as a pragmatic indicator of general psychopathological liability in childhood. This operationalization is supported by recent cross-national evidence indicating substantial convergence across different operationalizations of youth general psychopathology, including total problem scores and latent variable approaches (Achenbach et al., 2024), as well as by the absence of clear evidence supporting the superiority of more complex psychometric models in all contexts (Caspi et al., 2023; Levin-Aspenson et al., 2020). Nevertheless, the present operationalization should be interpreted as conceptually related to, but not psychometrically equivalent to, latent variable models of the p factor. The internal consistency of the scores obtained in our sample was excellent, with a Cronbach’s alpha coefficient of 0.94.
Emotion regulationEmotion regulation strategies were evaluated using the Emotion Regulation Checklist (ERC) by Shields and Cicchetti (1997). This parent-reported questionnaire comprises 24 items rated on a 4-point Likert scale (1 = rarely/never true to 4 = almost always true) that assess two dimensions: (1) emotional lability/negativity, reflecting emotional instability and difficulties in managing emotions, and (2) adaptive emotion regulation, which evaluates the ability to regulate and express emotions adaptively. The Spanish version of the ERC, used in previous studies (e.g. García & Míguez, 2007; Lucas-Molina et al., 2020), was used in this research.
In our sample, the internal consistency coefficients showed acceptable to good reliability, with Cronbach's alpha values of 0.63 for the Emotion Regulation subscale and 0.81 for the Lability/Negativity subscale.
Exposure to adverse childhood experiencesExposure to ACEs was assessed using the Center for Youth Wellness Adverse Childhood Experiences Questionnaire, CYW ACE-Q Child (Melnyk & Lusk, 2021). This parent-reported instrument assesses exposure to multiple forms of adversity occurring since birth, including family disruption, household dysfunction, interpersonal and community violence, and discrimination. In this study, two items assessing physical and sexual abuse were excluded from the original ACE-Q Child due to reliance on caregiver report and ethical considerations related to sensitive content.
Following the procedure used by other researchers (e.g. Nelson et al. 2021; Roman-Juan et al., 2023), we modified the original questionnaire, and the responses were categorized as binary (Yes/No). A total ACE score was calculated by summing affirmative responses, reflecting the cumulative number of adversities. The study utilized the Spanish version of the CYW ACE-Q Child, available online at https://centerforyouthwellness.org/aceq-pdf
Additionally, an ad hoc questionnaire was developed to collect sociodemographic information, including parental education, occupation, household income, family structure, and indicators of family psychopathology. This instrument was used to characterize the sample and provide contextual information about participants’ family background.
TemperamentChildren’s temperament was assessed using the Temperament in Middle Childhood Questionnaire (TMCQ), developed by Simonds and Rothbart (2004). This study utilized the Spanish translation by Barcons et al. (Universitat Autònoma de Barcelona) available at https://research.bowdoin.edu/rothbart-temperament-questionnaires/
The TMCQ is specifically designed to evaluate temperament dimensions in school-aged children. It includes 157 items rated on a 5-point Likert scale, measuring 17 specific temperament dimensions. These dimensions are grouped into four higher-order factors (negative affectivity, effortful control, surgency, and positive affectivity) that have been supported by prior research (e.g. Del Puerto-Golzarri et al., 2022).
For our sample, internal consistency was adequate to good for most scales, with Cronbach’s alpha coefficients of 0.88 for Effortful Control, 0.87 for Surgency, 0.66 for Behavioural Inhibition and 0.68 for Negative Affectivity scale.
ProcedureThis is a cross-sectional, observational study using parent-reported data. All variables (exposure, and outcome) were collected at the same time point as part of a broader child development research project between September 2024 and May 2025.
Parents reported their children’s behaviour using a custom-designed mobile application developed for research purposes, GroWinAPP (Universitat de les Illes Balears, 2024a, 2024b). Participants were recruited through multiple strategies, including the project’s web platform, the regional Department of Education, collaborations with schools, and engagement with parents’ associations. Recruitment efforts also extended to the general population via posters, press announcements, and news articles in local media. Once enrolled, parents accessed the app and were presented with all the study questionnaires, which they were asked to complete within 24–48 h.
Test administration was standardized across participants through the GroWinAPP platform. All caregivers received the same written instructions from the standardized questionnaires, completed the same questionnaire battery, and responded using identical item presentation and response formats. The app did not allow participants to skip individual items within each questionnaire, thereby minimizing item-level missing data and administration errors. Families who did not complete the full questionnaire battery were excluded from the analyses using the same criterion for all participants. Caregivers could contact the research team to resolve procedural doubts.
Statistical analysisAnalyses were conducted using complete cases for the variables included in the path model. The initial dataset included 328 child-level records; of these, 241 had complete data for all study variables and constituted the final analytic sample. The remaining 87 records were excluded because the full questionnaire battery was not completed. Because the GroWinAPP platform did not allow item-level omissions within each questionnaire, missingness was mainly due to incomplete questionnaire batteries rather than skipped individual items. Data were screened for potential out-of-range values prior to analysis. No cases were excluded as outliers; all available complete cases meeting the inclusion criteria were retained for analysis.
Preliminary descriptive statistics (means, standard deviations, ranges) were computed to characterise the sample. A path analysis was conducted to examine the associations between cumulative adverse childhood experiences (ACEs), specific temperament dimensions (negative affectivity, effortful control, behavioural inhibition, and surgency), emotion regulation processes (emotional lability and adaptive regulation), and the general psychopathology factor (p factor).
All variables were modelled as observed indicators. Cumulative ACEs were specified as an observed count score. Temperament dimensions were entered simultaneously as exogenous predictors. Emotional lability and adaptive regulation were specified as intermediary endogenous variables, and the p factor was modelled as the final endogenous outcome variable. Age and sex were included as covariates in all regression paths.
Model parameters were estimated using the robust weighted least squares estimator (WLSMV). Indirect effects were estimated using the product-of-coefficients approach, and their standard errors were obtained via bootstrapping. Statistical analyses were conducted in R (Version 4.4) using the Lavaan package (Rosseel, 2019).
ResultsDescriptive statistics and adversity exposureA summary of the sociodemographic characteristics of participating families is provided in Supplementary Table S1. Most caregivers were married and cohabiting with a partner (63.4% in first marriage; 80.7% currently living with a partner), while smaller proportions were single (15.9%) or separated/divorced (8.6%).
The sample was characterized by relatively high educational attainment: nearly half of caregivers held a university degree (45.9%), and an additional 22.3% reported postgraduate education (master’s or PhD). Occupational status was broadly distributed, with 35.2% employed in scientific or intellectual professions and 17.2% in executive or management positions. Unemployment rates were low (2.6%).
Household income was moderately distributed across categories. Approximately one-third of parents reported a lifetime mental health diagnosis (32.6%) or prior psychotropic medication use (33.5%), and 20.6% reported a history of mental health–related medical leave.
Children in the sample were exposed to an average of 0.75 adverse events (SD = 1.38), reflecting the mean number of Adverse Childhood Experiences (ACEs) reported, with individual scores ranging from 0 to 11. The most frequently endorsed ACEs were household mental illness or suicide attempt (18.3%) and parental separation or divorce (15.3%). These were followed by household substance abuse (7.9%), emotional abuse or threat (7.0%), and bullying (6.6%). All other adversities—such as exposure to community violence, recurrent unmet basic needs, or discrimination—were reported at very low frequencies. Further details on less commonly reported adversities are provided in Supplementary Table S2.
Model overview and explained varianceModel interpretation focused on the magnitude and statistical significance of structural paths and the proportion of variance explained in the endogenous variables.
Based on adjusted estimates, the model explained 21.3% of the variance in emotional lability and 45.1% of the variance in the p factor.
Direct associations among adversity, temperament, emotion regulation, and the p factorFig. 1 displays the standardized direct effects and their statistical significance (see Supplementary Table S3 for complete results). Cumulative adversity was positively associated with emotional lability, whereas its association with adaptive emotion regulation (AER) was non-significant.
Path analysis examining direct associations among cumulative adversity, temperament dimensions, emotion regulation processes, and the p factor.
Temperament dimensions showed differential associations with emotional lability. Negative affectivity, behavioural inhibition, and surgency were positively related to lability, whereas effortful control was negatively associated. None of the temperament dimensions were significantly associated with adaptive regulation in the path model.
Emotional lability demonstrated a strong and statistically significant direct association with the p factor. In contrast, adaptive emotion regulation was not significantly associated with the p factor once all other variables in the model were included.
Regarding direct effects on the p factor, cumulative adversity and behavioural inhibition showed significant positive associations. Effortful control showed a marginal negative association, and surgency demonstrated a small negative association. Negative affectivity did not retain a significant direct effect when mediators were included.
Indirect and total associations through emotion regulation processesStandardized indirect, and total effects on the p factor are summarized in Table 1; complete results are reported in Supplementary Table S4.
Standardized Indirect, and Total Effects on the p factor.
Note. Coefficients are standardized (β). * p < .05, ** p < .01, *** p < .001.
Significant indirect effects via emotional lability were observed for cumulative adversity and all temperament dimensions.
Notably, negative affectivity and effortful control showed significant indirect associations through emotional lability despite the absence of significant direct associations with the p factor. Behavioural inhibition showed both direct and indirect associations, as its association with the p factor was supported by both a direct path and an indirect path through emotional lability.
Surgency demonstrated a small indirect effect via lability; however, its total effect was not statistically significant.
Indirect effects through adaptive emotion regulation were not supported.
Overall, emotional lability emerged as the main intermediary correlate in the associations between cumulative adversity and temperament dimensions to p factor.
Covariate effectsAge and sex were included as covariates in all structural equations. Age showed a significant negative association with emotional lability (β = −.134, p = .009), indicating lower levels of lability at older ages. However, age was not significantly associated with adaptive emotion regulation (β = −.060, p = .290) or with the p factor (β = −.016, p = .704) after accounting for other exogenous variables in the model.
Sex was not significantly associated with emotional lability (β = 0.027, p = .664), adaptive emotion regulation (β = 0.104, p = .111), or the p factor (β = 0.057, p = .214).
Overall, age showed a modest association with emotional lability, whereas sex did not demonstrate significant contributions to the main endogenous variables once cumulative adversity and temperament dimensions were included.
DiscussionBy modelling dispositional and contextual risk factors within a single structural framework, the findings provide an integrative account of the associations between contextual and dispositional risk factors and broad psychopathological liability during this developmental period. Specifically, the present study examined how adverse childhood experiences, temperament dimensions, and emotion regulation processes jointly contribute to general psychopathological vulnerability in middle childhood.
The model explained a substantial proportion of variance in the p factor (45%), with emotional lability emerging as the strongest and most consistent correlate. Both temperament and cumulative adversity showed meaningful associations with general psychopathology, operating through distinct but partially overlapping pathways.
Across analyses, emotional lability demonstrated the most robust direct association with the p factor. This finding reinforces theoretical models that conceptualize heightened emotional reactivity and instability as core processes underlying transdiagnostic vulnerability in childhood. Emotional lability reflects rapid, intense, and poorly modulated affective responses, which may be closely related to broader psychopathological liability. These findings may also be interpreted within developmental models conceptualizing temperament as a biologically based vulnerability system involved in emotional responding under stress. In line with the Regulative Theory of Temperament, heightened emotional reactivity may predispose children to stronger emotional responses and greater difficulties regulating emotional activation following adverse experiences (Strelau, 1998; Strelau & Zawadzki, 2005). From this perspective, emotional dysregulation may represent an intermediary emotional process linking dispositional vulnerability and exposure to adversity with broader psychopathological liability.
In contrast, adaptive emotion regulation (AER) accounted for minimal explained variance and did not show an independent association with general psychopathology when controlling for lability and the other variables included in the model. This distinction suggests that, during middle childhood, vulnerability may be more closely associated with dysregulated emotional responses than by limited access to adaptive regulatory strategies. This pattern is consistent with prior evidence indicating that maladaptive or dysregulated emotional processes exert stronger associations with mental health outcomes than the mere absence of adaptive regulation (e.g., Aldao et al., 2010; Compas et al., 2017; Weissman et al., 2019).
Taken together, these findings indicate that dysregulated emotional reactivity may constitute a more proximal correlate of general psychopathology than adaptive regulatory skills, at least within community samples in this developmental period.
This distinction is theoretically relevant because emotional lability and adaptive regulation may not represent opposite ends of the same continuum. Emotional lability captures overt affective instability, rapid escalation, and poorly modulated emotional responses, which are likely to be highly visible to caregivers and closely intertwined with broad behavioural and emotional difficulties. By contrast, adaptive regulation may reflect a more heterogeneous set of developing competencies, including emotional understanding, socially appropriate expression, and the capacity to use regulatory strategies when needed. These skills may support adjustment, but their association with broad psychopathological liability may be weaker or more conditional once dysregulated emotional reactivity is taken into account. The minimal variance explained in adaptive regulation also highlights potential conceptual and measurement considerations, as adaptive regulation may constitute a more voluntary, dynamic, and context-dependent process than emotional lability. Rather than reflecting a fixed protective capacity, adaptive regulation may depend on learning history, situational demands, and the functional appropriateness of specific strategies. There are arguably no universally “adaptive” or “maladaptive” strategies a priori; rather, regulatory effectiveness depends on contextual fit and flexibility (e.g., Aldao et al., 2015; Kobylińska & Kusev, 2019).
Cumulative adverse childhood experiences demonstrated both direct and indirect associations with the p factor. Specifically, adversity was linked to higher emotional lability, which in turn was associated with greater general psychopathology. In addition, a direct association between adversity and the p factor remained after accounting for temperament and regulation processes. These findings indicate that contextual adversity contributes to broad psychopathological vulnerability both through heightened emotional instability and through additional pathways not fully captured by the present model.
This pattern is consistent with research linking early adversity to broad and transdiagnostic mental health outcomes (e.g., Kirkbride et al., 2024) and with evidence that adversity exposure may heighten emotional reactivity, thereby increasing vulnerability to broad psychopathology.
Although ACEs have traditionally been conceptualized as severe experiences concentrated on highly vulnerable populations, more recent research has emphasized the importance of broadening the definition of adversity to include a wider range of contextual stressors (Hawes & Allen, 2023). In line with this perspective, the present findings indicate that even relatively common home-context adversities—such as parental mental health difficulties or parental separation—may carry meaningful implications for child psychopathology.
In this community sample, cumulative adversity levels were generally low to moderate and primarily reflected everyday family stressors rather than severe maltreatment. Nevertheless, cumulative exposure was significantly associated with general psychopathology in middle childhood. Even though the magnitude of these effects was more modest than that of emotional lability, contextual risk exerted measurable influence even within a relatively socioeconomically stable population.
These findings support a dimensional view of adversity, whereby risk operates along a continuum and is not restricted to contexts of extreme deprivation. The emergence of significant associations during primary school years further suggests that the impact of cumulative adversity may become evident early in development, underscoring the importance of prevention efforts that extend beyond traditionally defined high-risk groups.
Temperament dimensions demonstrated differentiated patterns of association with the p factor. Behavioural inhibition and effortful control emerged as the traits with the strongest overall contributions, while operating in opposite directions. Behavioural inhibition showed a robust positive total association with general psychopathological vulnerability, operating through both direct and lability-mediated pathways. In contrast, effortful control demonstrated a comparably sized protective association, with lower regulatory control linked to higher p factor scores primarily through increased emotional lability. The near equivalence in magnitude between these two traits underscores the central role of regulatory control and threat sensitivity in shaping transdiagnostic liability.
Negative affectivity, although substantially associated with the p factor, operated exclusively through emotional lability, indicating that dispositional negative emotionality may confer vulnerability primarily by intensifying emotional instability and dysregulation rather than through independent pathways. From a hierarchical perspective, negative affectivity has been identified as the temperament trait most consistently and robustly linked to psychopathology, including the p factor (e.g., Brandes & Tackett, 2019). The present findings do not contradict this evidence but rather refine it, suggesting that the association between negative affectivity and the p factor may be largely accounted for by intermediary dysregulatory processes. In this sense, broad negative emotionality may function as a distal liability dimension whose influence on transdiagnostic vulnerability becomes manifest when translated into emotional instability.
This interpretation may help clarify mixed longitudinal evidence regarding the predictive role of negative affectivity. For instance, Morales et al. (2021) did not observe consistent direct associations between early negative affectivity and later general psychopathology. Our results suggest that such inconsistencies may arise when intermediary regulatory mechanisms are not explicitly modelled, as the impact of dispositional negativity appears to operate through emotional lability.
Conceptually, these findings support a distinction between distal dispositional vulnerability and proximal dysregulatory processes more directly implicated in the p factor (Lahey et al., 2017; Wu et al., 2022).
Surgency displayed a differentiated pattern of associations. Although its direct effect on the p factor was negative—suggesting a potential protective role—surgency also showed a small positive indirect association via emotional lability. This configuration indicates that surgency may confer both protective and risk-related influences, depending on how approach-related tendencies interact with emotional reactivity. In other words, while aspects of social engagement and positive affect may attenuate broad psychopathological vulnerability, heightened approach motivation may also be linked to greater emotional instability in certain contexts. However, given the non-significant total effect, these interpretations should be considered exploratory and hypothesis-generating rather than conclusive.
Overall, these findings indicate that dispositional traits contribute to general psychopathological vulnerability through partially overlapping but distinct pathways.
Emotional lability emerges as a central intermediary process linking dispositional traits—particularly negative affectivity, effortful control, and behavioural inhibition—to the broader architecture of the p factor.
Limitations and future directionsDespite the contributions of this study, several limitations must be acknowledged that may affect the interpretation and generalizability of the findings.
First, the cross-sectional design prevents establishing causal relationships between the variables under study. Although the proposed model is theoretically grounded and supported by previous empirical evidence, longitudinal research is essential to confirm the directionality of effects and to trace the developmental trajectory of the p factor over time.
Second, the sample size was relatively modest, although it represents a valuable community sample. In addition, the limited representation of socioeconomically disadvantaged and culturally diverse families may restrict the generalizability of the findings. Because participants were drawn from a specific sociocultural context, future research should examine whether these results generalize to more socioeconomically and culturally diverse populations.
Additionally, the use of a cumulative ACE index may obscure differential effects of timing, severity, or specific types of adversity. Although this approach captures the overall burden of exposure to environmental risk, it does not allow us to examine the contribution of distinct adversity dimensions or broader contextual processes, such as parenting practices, family functioning, socioeconomic stress, or school and community factors.
Third, all measures were based on caregiver report. In middle childhood—and particularly among younger children—parental reports constitute a primary and developmentally appropriate source of information regarding children’s emotional and behavioural functioning, as well as family-context adversities. Nevertheless, reliance on a single informant and the absence of more objective measures may introduce shared method variance or perceptual biases.
This issue is particularly relevant when interpreting the association between emotional lability and broad psychopathology. Although emotional lability and the CBCL Total Problems Score are conceptually related, they are not equivalent constructs. Emotional lability reflects affective instability and difficulties modulating emotional responses, whereas the CBCL Total Problems Score indexes a broader and more heterogeneous range of psychopathological manifestations, including internalizing, externalizing, attentional, social, somatic, thought-related, and behavioural problems. However, some conceptual and empirical overlap may exist because both measures include aspects related to emotional and behavioural dysregulation, and this overlap may have been amplified by the use of a single informant. Therefore, the strength of the association between emotional lability and p factor should be interpreted cautiously. Additionally, the operationalization of the p factor relied on an observed dimensional proxy rather than a latent bifactor model. Although this approach is supported by evidence showing convergence across operationalizations of broad psychopathology, the findings should be interpreted within the constraints of this psychometric approach. Future studies would benefit from incorporating multi-informant approaches (e.g., teacher reports, clinical observations, child self-reports, or other objective indicators), as well as latent modelling approaches, to strengthen the robustness and external validity of findings.
Finally, questionnaire-based measures may not fully capture the contextual and dynamic nature of regulatory processes. Because adaptive emotion regulation often depends on situational demands and strategy–context fit, single-time self-report or caregiver-report instruments may provide only a limited representation of these processes. Future research would benefit from using more ecologically valid methodologies, including real-time and naturalistic assessment approaches, to better capture regulatory flexibility, contextual sensitivity, and the dynamic nature of emotional functioning across development.
ConclusionIn a community sample of children aged 6–12 years, the present path model integrates dispositional and contextual risk factors, representing an attempt to examine the interplay between individual and environmental vulnerability in the emergence of broad psychopathological liability. Specifically, the model integrates cumulative adversity, temperament, and emotion regulation processes to clarify how the general psychopathology factor is structured during this developmental period.
Within the model, emotional dysregulation emerged as the most proximal and robust correlate of the p factor, accounting for substantial indirect associations linking both adversity and dispositional traits to general psychopathology. Temperament showed differentiated contributions: behavioural inhibition and effortful control carried the strongest total associations (in opposite directions), negative affectivity related to the p factor indirectly via lability, and surgency showed a small counterbalancing pattern with a non-significant total effect. Cumulative adversity retained both direct and indirect associations through emotional lability, suggesting that common family-context stressors may be related to broad vulnerability through emotional instability, as well as through additional processes not captured by the present model.
Notably, adaptive regulation showed minimal explained variance and no independent association with the p factor, underscoring the possibility that, at this developmental stage, generalized vulnerability is more tightly coupled to dysregulated affective reactivity than to reported adaptive skills.
From a prevention perspective, and in line with transdiagnostic models of psychopathology and recent calls for earlier, developmentally informed approaches to youth mental health (McGorry et al., 2024), the present findings underscore the importance of identifying modifiable intermediary processes. Because temperament and adversity are relatively stable vulnerability dimensions, focusing on emotional lability may represent a promising strategy for early interventions aimed at reducing general psychopathology.
FundingProyecto PID2021-126704OB-I00 financiado por MCIN/AEI /10.13039/501100011033 y por FEDER, UE.
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
The authors would like to thank all participating children, families, and teachers for their collaboration in this study and for their support during data collection.
