Exposure to Parental Aggression and the Development of Psychopathology in Young Children: The Mediating Role of Early Dissociative Symptoms

Abstract

Exposure to child-directed parental aggression in early life has been found to increase the risk of later psychopathological symptoms among children and adolescents. However, little is known about intermediate phenotypes and the developmental progression of symptoms, especially across the transition to grade school. Using prospectively collected longitudinal data from a large sample of children enrolled in the Family Life Project (n = 1,166, 49.7% female), the current study examined the mediating role of early dissociative symptoms in the relations between parental aggression and children’s psychopathological symptoms. Children’s exposure to parental aggression and their dissociative symptoms before school entry were assessed based on primary caregivers’ reports. Teacher ratings of children’s internalizing and externalizing symptoms were collected in pre-kindergarten as well as in the 1st, 3rd, and 5th grades. Results showed that dissociative symptoms before school entry partially mediated the association between parental aggression and persistent externalizing symptoms in school years. However, no significant associations were found between parental aggression or dissociative symptoms and internalizing symptoms. Findings suggest that dissociative symptoms manifested early in life serve as a mediating mechanism and indicator of risk for persistent impulsivity and behavioral problems. Thus, these symptoms could be an important target of preventive services provided to children with adverse experiences in their families.

Keywords

parental aggression dissociative symptoms internalizing and externalizing symptoms developmental mechanism structural equation modeling Family Life Project

Introduction

Although family is usually the primary source of care and protection for young children, many children also experience aggression directed toward them by their caregivers. According to Child Protective Services and community professionals’ reports on cases involving actual harm or substantial endangerment, approximately 476,600 children in the United States experienced documented physical abuse and 302,600 children experienced psychological abuse from caregivers in the year of 2005–2006 (Sedlak et al., 2010). These numbers are considered to be an underestimate of parental aggression, as they do not account for undetected/unreported incidents, or less severe aggression that may not be deemed as abuse. A more accurate indication of the prevalence of parental aggression may be taken from surveys of parents, which suggest that approximately 70% of toddler- and preschool-aged children in the United States experience at least some physical and/or psychological aggression from parents, with as many as 10–20% exposed to severe parental aggression (e.g., hitting or kicking hard, cursing at child; Runyan et al., 2010; Slep & O’Leary, 2005).

Parents’ physical and psychological aggression toward children, ranging from relatively mild corporal punishment and verbal attacks, to more severe physical and psychological abuse, has been related to increased risk of developing psychopathological symptoms on both internalizing and externalizing spectrums (Gershoff, 2002; Melançon & Gagné, 2011; Riina et al., 2014). Although exposure to parental aggression often starts early in life, related negative consequences are likely to manifest over time. In particular, when children start formal schooling and face greater demands for demonstrating socially adaptive emotions and behaviors, an underlying regulatory deficiency may start to manifest as more salient behavioral and emotional problems (Denham et al., 2014; Pears et al., 2010), leading to a range of negative developmental outcomes and increased needs for clinical and social services (Atkins et al., 2017; Jonson-Reid et al., 2007). Due to the comorbidity and overlap in behavioral manifestations between trauma symptoms and internalizing or externalizing symptoms among school-age children and adolescents (Ford et al., 2000; Linning & Kearney, 2004), symptoms and developmental mechanisms directly associated with exposure to parental aggression may often be overlooked, impacting the precision of diagnosis and treatment (Szymanski et al., 2011).

To understand the association between parental aggression and psychopathological symptoms, and to provide prevention for at-risk children at early ages, it is critical to identify mediating mechanisms and early indicators of risk. However, due to a lack of prospective longitudinal studies and measurement of potential mediators, there has been limited evidence on the developmental progression of symptoms and intermediate phenotypes. Potential mediating pathways may exist at different levels of analysis. For instance, social modeling theories (e.g., Bandura, 1978) propose that children exposed to parental aggression learn to mimic these behaviors. At the biological level, chronic exposure to unpredictable stress and violations of a sense of security has been shown to alter neuroendocrine functioning that may increase vulnerability to both internalizing and externalizing psychopathology (Danese & McEwen, 2012). However, one particularly salient mechanism may be the tendency to develop coping responses in the context of acute threat that do not align with adaptive coping and regulatory skills needed in typical social and interpersonal contexts (Wadsworth, 2015).

One set of symptoms associated with such coping responses is dissociation, which refers to a lack of consistent access to and integration of concurrent thoughts, feelings, perceptions of situational contexts, and memories (Putnam, 2006). In severe and relatively rare cases, dissociation can manifest as clinical disorders (e.g., pathological amnesia, identity disorders; American Psychiatric Association, 2013) in its own right. However, dissociative symptoms (e.g., memory lapses, trance-like states, unpredictable shifts in mood and knowledge) can exist more commonly on a continuum and are more salient among children exposed to parental aggression at an early age (Hulette et al., 2011; Vonderlin et al., 2018). Several theoretical models have been proposed to account for the developmental mechanisms of dissociation, and many agree that it reflects a tendency to avoid or disconnect from overwhelming distress related to threats and/or traumatic experiences (Dalenberg et al., 2012; Silberg, 2014). Models drawing on attachment theories also suggest that frightening and unpredictable parental behaviors hinder children’s development of integrated internal working models to cope with distress, placing them at higher risk for developing dissociative symptoms (Lyons-Ruth et al., 2006). Because young children have very limited options for coping with the often unavoidable threats imposed by parental aggression, they may develop a reliance on dissociation that offers relief from acute distress in the moment (Putnam, 2006). Dissociation may thus be reinforced and generalized over time as a primary response when facing stressors, which can interfere with the development of coping strategies with greater long-term adaptive value.

Although most children do not demonstrate clinically significant dissociation, symptoms in this domain can impact the growth of self-regulation and socialized behaviors that characterize normative development in early childhood. Dissociative symptoms have been related to risk for both internalizing and externalizing symptoms (Ensink et al., 2017; Silvern & Griese, 2012), and such risk may unfold through multiple pathways. For instance, one of the developmental tasks in early childhood is to grow awareness of and be able to differentiate one’s own emotional states. These abilities further support the regulation of emotions, which serves a protective role against developing psychopathological symptoms (Stegge & Terwogt, 2007). When children rely on dissociative behaviors and generalize such responses to cope with daily stressors, they may have less opportunity to develop emotional awareness. Furthermore, they may be less efficient in acquiring more cognitively advanced coping strategies that promote adaptive socioemotional skills (e.g., cognitive reframing). This may then result in greater probability of developing both internalizing and externalizing symptoms. Across early childhood, children are also expected to develop the ability to voluntarily control attention and behaviors and to incorporate social expectations for behaviors and emotions through daily experience (Calkins, 2007). Children who rely on dissociative behaviors may have difficulties integrating perceptions of their own states and their surroundings (Silberg, 2014), which can prevent them from learning social expectations, developing empathy, and practicing self-control. Dissociative behaviors may thus lead to greater risk for externalizing problems including aggressive behaviors and impulsivity.

To summarize, dissociation may be one of the mediating mechanisms through which parental aggression leads to risk for psychopathological symptoms. Understanding this mechanism can help inform efforts to prevent negative developmental outcomes among children with adverse early experiences. Specifically, parental aggression often starts early in life, but related adjustment difficulties may not become salient until school years when at-risk children fail to keep up with the developmental demands in academic and social contexts. Elevated dissociative behaviors in reaction to the experience of violence, however, may emerge early, and thus can be an indicator of risk for further adjustment difficulties or a potential target of prevention/intervention.

Previous studies examining the mediating role of dissociative symptoms in the association between childhood traumatic experience and later maladjustment have typically focused on adult and adolescent clinical samples and relied on retrospective recall. In addition to potential biases related to the retrospective approach, studies with adult and adolescent samples cannot identify the developmental course of the risk mechanisms among young children, and may therefore be less informative for preventive efforts (e.g., how to best identify children at risk at an early stage). To address this gap, the current study utilized prospectively collected longitudinal data to test the following hypotheses: (a) experience of parental aggression in early life would be associated with elevated dissociative symptoms manifested before school entry, and (b) early dissociative symptoms would mediate further risk for developing internalizing and externalizing problems in later school years.

Methods

Participants

This study used data drawn from the Family Life Project, a prospective longitudinal study of 1,292 children and their families. The sample was recruited from six rural counties in North Carolina and Pennsylvania, which were characterized by high rates of poverty. Families were recruited from local hospitals at the time of the child’s birth between September 1, 2003, and August 31, 2004. Low-income families were oversampled in both states, and African American families were oversampled in the North Carolina counties. Over the ensuing decade, participating families completed assessments in a series of home visits. When children started attending schools, their teachers were asked to provide ratings on children’s behaviors and adjustment. Complete details of sampling structure and study procedures can be found in Vernon-Feagans et al., (2013). All procedures were approved by the Institutional Review Board at the University of North Carolina (with reliance from the Pennsylvania State University). Parents provided written consents for their participation and for their child’s participation, as well as permission to contact the child’s teacher.

Among the enrolled participants, 126 who did not have available data on any of the main study variables (i.e., parental aggression, dissociative symptoms, and teacher reports on psychopathological symptoms) were excluded from the current study. There were no significant differences between the excluded sample and the analysis sample with regard to children’s sex or race, primary caregivers’ education, or family income-to-needs ratio. Analyses in the current study included n = 1,166 children (49.7% female). Slightly over half of the analysis sample identified their child as Caucasian (55.8%), with the remaining identifying their child as African American (42.7%), or other races (1.5%). At the time of the child’s birth, 64.4% of the families had annual incomes ≤200% of the federal poverty threshold (i.e., very poor or nearly poor). Among the primary caregivers (99.6% were biological mothers, and .4% were grandmothers or other female relatives; M_age = 25.92 years at childbirth, SD = 6.07), 24.3% did not have a high school diploma, 61.3% had a high school diploma but no secondary degree, and 14.4% held a college degree.

Measures

Parental aggression toward child.

During the home visit when the child was approximately 36-months old, the primary caregiver filled out a questionnaire adapted from the Parent–Child Conflict Tactics Scale (CTSPC; Straus et al., 1998). This questionnaire listed a series of behaviors adults might engage in when having a conflict/disagreement with the child or getting annoyed with the child. The primary caregiver self-reported how often they engaged in each behavior on a 7-point scale (0 = Never, 1 = Less than once a month, 2 = Once a month, 3 = 2–3 times a month, 4 = Once a week, 5 = 2–3 times a week, 6 = Almost every day). If the primary caregiver had a spouse/partner at the time (74.0% of the sample, regardless of whether the partner lived in the household), they also rated how often the partner engaged in each behavior toward the child during the past year.

To assess children’s exposure to parental aggression, this study focused on two subscales. The physical aggression subscale included 7 items describing corporal punishment and other violent physical acts toward the child (e.g., “Spanked your child with something;” “Pushed, grabbed, or shoved your child;” “Hit or tried to hit your child”). The psychological aggression subscale consisted 8 items describing parents’ verbal or non-verbal expressions of humiliation, rejection, intimidation, or threats (e.g., “Yelled, insulted, or swore at your child;” “Stomped out of the room or house;” “Threatened to beat up your child”).

The internal consistency was moderate for individual physical or psychological aggression subscale (Cronbach’s αs = .54–.69), potentially due to the relatively low occurrence rates of some behaviors. However, the internal consistency was satisfactory across items from both subscales (Cronbach’s α = .80 for primary caregivers’ self-report, and .72 for their report on the partner’s behaviors). Scores were averaged across items respectively for the physical and psychological aggression subscales, as well as for total aggression (including items from both subscales). The average scores reflected the presence and chronicity of aggressive behaviors (Straus et al., 1998). Given existing evidence that aggression from multiple caregivers may have additive effects in predicting child outcomes (e.g., Lee et al., 2013), the subscale/total aggression scores were summed across the primary caregiver and the spouse/partner for each child (i.e., if there was no partner, it was assumed that the child has no opportunity to be exposed to aggression from this person, so the final score was then equal to the primary caregiver’s score). Thus, the possible range of the final physical, psychological, or total aggression score is 0–12. As reported in Table 1, the observed range for total parental aggression was 0–8, indicating the presence of fairly high aggression, although the mean level of aggression was M = 0.72.

Table 1.

Means, Standard Deviations, Ranges, and Bivariate Correlations of Study Variables.

Variable	1	2	3	4	5	6	7	8	9	10
1. Physical aggression–36m	–
2. Psychological aggression–36m	.67*	–
3. Total parental aggression–36m	.90*	.93*	–
4. Dissociative symptoms–58m	.19*	.16*	.19*	–
5. INT–PK	.06	.05	.06	.11*	–
6. CP–PK	.18*	.10*	.15*	.17*	.16*	–
7. HY–PK	.14*	.07	.11*	.23*	.23*	.65*	–
8. INT–school years	.01	.04	.03	.07*	.16*	.12*	.17*	–
9. CP–school years	.17*	.14*	.17*	.24*	.04	.46*	.40*	.28*	–
10. HY–school years	.18*	.13*	.16*	.22*	.07	.37*	.52*	.34*	.67*	–
Mean	0.64	0.78	0.72	4.97	0.23	0.23	0.67	0.31	0.30	0.75
SD	0.86	0.88	0.80	5.98	0.33	0.37	0.58	0.32	0.36	0.55
Observed minimum	.00	.00	.00	.00	.00	.00	.00	.00	.00	.00
Observed maximum	8.00	8.00	8.00	40.00	1.80	2.00	2.00	1.67	1.80	2.00

Note. *p < .05.

36m/58m = Variable was obtained from parent report at the 36-month/58-month assessments.

PK= Teacher ratings in pre-Kindergarten; School years = Average of teacher ratings in the 1st, 3rd, and 5th grades. Total parental aggression was calculated across all physical and psychological aggression items.

INT = Internalizing symptoms; CP = Conduct problems; HY = Hyperactivity.

SD = Standard deviation; observed minimum/maximum reflect the actual ranges in the current sample.

Figure 1 further depicts the range and frequency of exposure to parental aggression. Based on primary caregivers’ report, the majority of the children were exposed to some aggression, with 29% and 41% respectively experiencing at least one item of physical or psychological aggression at least once a week; 5% and 11% of the sample were exposed to at least one item of physical or psychological aggression almost every day. Compared to the less-frequent exposure groups, caregivers in the more-frequent exposure groups were more likely to report engaging in behaviors described in multiple items (versus only a single item); χ² (1, n = 751) = 24.47, p < .001 for physical aggression, χ² (1, n = 831) = 42.86, p < .001 for psychological aggression. This suggests that children who were exposed to parental aggression more frequently also tended to experience a wider variety of aggressive behaviors.

Figure 1.

Note. The mean and range above each bar correspond to the physical or psychological aggression scores (averaged across subscale items and summed across primary caregiver and spouse/partner) of the specific group.

Child dissociative symptoms.

Children’s dissociative symptoms were assessed via the 20-item Child Dissociative Checklist (CDC; Putnam et al., 1993) during the home visit when the child was around 58 months’ old. The primary caregiver rated whether each item described the child’s behavior within the past 12 months on a 3-point Likert scale (0 = Not true, 1 = Somewhat true, and 2 = Very true). Sample items included “Child does not remember or denies traumatic or painful experiences that are known to have occurred,” “Child goes into a daze or trance-like state at times or often appears ‘spaced out’.” For each child, a sum score of all items was computed (possible range = 0–40). CDC has demonstrated good internal consistency and test-retest reliability in both community and maltreated samples of young children (Ensink et al., 2017; Putnam et al., 1993). The Cronbach’s α was .90 in the current sample. Previous research has established the validity of this scale, with scores of 12 or greater found to distinguish 96% of preschool- and school-aged children with a clinical diagnosis of dissociative disorders (Putnam & Peterson, 1994). The descriptive statistics are included in Table 1, with the distribution presented in Appendix A. Children’s scores spread the full range of the scale, and 12.5% (n = 133) of the sample who had available data on this scale scored within the clinical range (scores ≥ 12).

Child internalizing and externalizing symptoms.

Teachers were asked to complete the Strengths and Difficulties Questionnaires (SDQ; Goodman, 1997) when the child was in pre-kindergarten (pre-K), 1st, 3rd, and 5th grade. Internalizing symptoms were measured through the emotional symptoms subscale, which included 5 items describing somatic complaints (e.g., “Often complains of headaches, stomachaches or sickness”), fears and worries (e.g., “Many fears, easily scared”), and depressive symptoms (e.g., “Often unhappy, depressed, or tearful”). Another two 5-item subscales respectively assessed conduct problems (i.e., aggressive/oppositional behaviors, e.g., “Often fights with other children and bullies them,” and antisocial behaviors, e.g., “Often lies or cheats”) and hyperactivity (i.e., impulsive/hyperactive behaviors, e.g., “Restless, overactive, cannot stay still for long,” and inattentiveness, e.g., “Easily distracted, concentration wanders”). Each item was rated on a 3-point Likert scale ranging from 0 (not true) to 2 (certainly true), and the average of items in each subscale was computed for each year. All three subscales demonstrated good internal consistency with the current sample at each of the four assessment years (Cronbach’s αs = .70–.76 for emotional symptoms, .79–.83 for conduct problems, and .87–.89 for hyperactivity).

The pre-K measures were collected in the same year as caregivers’ reports on dissociative symptoms. Thus, they were used as covariates in subsequent analyses in order to examine the unique predictive value of dissociative symptoms beyond concurrent psychopathological symptoms. For each subscale, an average score across the 1st-, 3rd-, and 5th-grade assessments was computed (scores of consecutive assessments were significantly correlated; r = .44–.53 for conduct problem, r = .51–.56 for hyperactivity, and r = .28–.30 for emotional symptoms); higher scores thus indicate greater symptom severity and chronicity during school years. In order to capture the shared risk underlying both types of externalizing symptoms, conduct problems and hyperactivity scores were used to model a latent externalizing symptoms factor in subsequent analyses.

A total of 29% of the analysis sample did not have pre-K data, either because the teacher refused to participate, or that the child did not attend preschool. Children with and without pre-K data demonstrated similar levels of psychopathological symptoms in the 1st grade (which only had a 7% missing rate); t (423) = –.04, p = .97 for conduct problems, t (414) = .07, p = .94 for hyperactivity, and t (398) = .22, p = .82 for internalizing symptoms. Thus, the missingness of pre-K psychopathological symptoms was not likely related to the levels of symptoms themselves (i.e., not likely to be missing-not-at-random). Among all the study variables, the missingness was only related to primary caregivers’ education, such that children of mothers who did not complete high school were more likely to be missing pre-K data. As we statistically controlled for primary caregivers’ education in subsequent analyses, and the maximum likelihood estimation applied in structural equation modeling could handle missing-at-random patterns (Enders, 2010), the higher rate of missing in pre-K data was not expected to influence the interpretability and generalizability of results.

Based on the cut-offs proposed by Goodman (1997) specific to teacher-reported SDQ, 6.1–10.0% of children in the current sample showed borderline or abnormal levels of internalizing symptoms (≥ 1.0) across the pre-K, 1st-, 3rd-, and 5th-grade assessments. The borderline/abnormal percentages ranged from 22.0% to 29.5% across the four assessments for hyperactivity (≥ 1.2), and ranged from 15.8% to 25.7% for conduct problems (≥.6). Overall, these scores suggest higher incidence of externalizing psychopathology relative to the rates of diagnosis among the general population of children and adolescence in the United States (Centers for Disease Control and Prevention, n.d.).

Demographic variables.

Primary caregivers reported the child’s sex and race, and their own education during the home visit when the child was 2 months’ old. At six time points before school entry (6-, 15-, 24-, 36-, 48-, and 58-month home visits), they also reported their household income. Income-to-needs ratio was computed by dividing the annual income by the federal poverty threshold corresponding to the family’s household size for each specific year, and an average score across the six-time points was calculated for each family. To accommodate outliers, averaged income-to-needs ratio scores that were higher than 5 (n = 40) were recoded as 5.

Analytic Strategies

Structural equation modeling was used to examine whether parental aggression predicted higher levels of dissociative symptoms before school entry and whether these early symptoms mediated risk for psychopathological symptoms in school years. Analyses were conducted in R, and structural equation models were fitted using the lavaan package (v. .5–23.1097; Rosseel, 2012). Full information maximum likelihood estimation was used, and Huber-White robust standard errors were obtained based on observed information matrix to minimize the impact of data nonnormality on the estimation of standard errors and standardized coefficients (Lai, 2018; Savalei, 2010). Model fits were evaluated based on the χ² fit statistics (scaled as equal to the Yuan-Bentler test statistic) and other indices including CFI, TLI, RMSEA, and SRMR (Bentler, 2007). Bootstrapping with 1,000 replications was used to obtain the 95% confidence intervals of all estimated effects (Falk, 2018).

The two outcome measures, internalizing and externalizing symptoms in school years, were examined in two separate mediational path models (refer to Figures 2a and 2b). Regression paths from parental aggression to dissociative symptoms (a path), and from dissociative symptoms to the outcome measure (b path), were assessed. The indirect effects (a*b) through dissociative symptoms were estimated, and the remaining direct effect of parental aggression on the outcome measure (c path) was also examined. An auto-regressive path was added between internalizing or externalizing symptoms in pre-K and those in school years, and the pre-K measure was regressed on parental aggression. A correlation path was also estimated between the pre-K measure and dissociative symptoms. Thus, the mediational effect (if any) of early dissociative symptoms would be unique beyond the predictive value of psychopathological symptoms also measured before school entry. Parental physical and psychological aggression scores were highly correlated (r = .67), and results were similar when they were evaluated in separate models. Thus, the total aggression score (including items in both subscales) was used as the predictor in the final models.

Figure 2.

Note. Standardized regression coefficients and factor loadings, and correlation coefficients among main study variables are displayed.

Child sex and race, primary caregiver’s education, and income-to-needs ratio were entered into the models as covariates. Primary caregivers’ education were coded into three categories: did not complete high school, completed high school but not college, and held a college degree. Two variables were then created based on simple effect contrast coding, using the second category as the reference. Thus, in the analyses, “below high school” and “college or above” respectively compared parents who did not complete high school and parents who had a college degree to those who completed high school but not college.

Results

We first evaluated the overall fits of the two mediational models. The model of internalizing symptoms demonstrated a good fit with the data, χ² (4) = 2.20, p = .70, CFI = 1.00, TLI = 1.02, RMSEA < .01, SRMR = .01. For the model predicting externalizing symptoms, although the absolute fit statistic indicated a statistically significant discrepancy with the data, χ² (17) = 54.75, p < .01, all other indices suggested satisfactory fit, CFI = .99, TLI = .96, RMSEA = .04, SRMR = .02. As the absolute fit statistic is overly sensitive to trivial discrepancies with large sample sizes (Fan et al., 1999), we did not reject the model solely based on the χ² test and proceeded to examining specific paths.

As suggested in both models (refer to Tables 2 and 3; covariances are presented in Appendix B), higher levels of parental aggression measured at 36-month were associated with more dissociative symptoms of children at 58-month (path as). Children’s race, but not sex, was related to the level of symptoms, such that minority children demonstrated more dissociative symptoms. Lower income-to-needs ratio and lower education level of the primary caregiver were also associated with more dissociative symptoms in children.

Table 2.

Parameter Estimates of Regression Paths in the Model of Internalizing Symptoms.

Dependent Variable	Parameter Estimates
Predictor	Coefficient (SE)	β	95% CI
Dissociative symptoms
Child sex	–.11 (.35)	–.01	[–.64, .46]
Income-to-needs ratio	–.58* (.17)	–.12	[–.85, –.31]
Race	1.05* (.40)	.09	[.40, 1.71]
Maternal education (below high school)	1.37* (.50)	.10	[.55, 2.16]
Maternal education (college or above)	–1.56* (.38)	–.09	[–2.16, –.95]
Total parental aggression (path a)	1.26* (.34)	.17	[.72, 1.86]
Internalizing symptoms in PK
Child sex	–.01 (.02)	–.02	[–.05, .03]
Income-to-needs ratio	–.02 (.01)	–.06	[–.04, .01]
Race	.02 (.03)	.03	[–.03, .06]
Maternal education (below high school)	.04 (.03)	.05	[–.01, .09]
Maternal education (college or above)	–.03 (.03)	–.03	[–.07, .03]
Total parental aggression	.02 (.02)	.05	[–.01, .05]
Internalizing symptoms in school years
Child sex	<.001 (.02)	<.001	[–.03, .03]
Income-to-needs ratio	–.02* (.01)	–.08	[–.04, –.004]
Race	–.03 (.02)	–.05	[–.07, .003]
Maternal education (below high school)	.02 (.03)	.03	[–.02, .07]
Maternal education (college or above)	–.07* (.03)	–.08	[–.12, –.02]
Internalizing symptoms in pre-K	.13* (.04)	.14	[.07, .20]
Dissociative symptoms (path b1)	.001 (.002)	.02	[–.002, .004]
Total parental aggression (path c1)	.006 (.01)	.01	[–.02, .03]

Note. *p < .05.

Coefficient = Unstandardized coefficient; SE = Standard error; β = Standardized coefficient; CI = Bootstrapped confidence interval.

Child sex was coded as 0 (female) and 1 (male). Race was dichotomously coded as 0 (Caucasian) and 1 (African American and others).

The two maternal education variables respectively reflect the effects of not completing high school and having completed college, when compared to primary caregivers who completed high school but not college.

Table 3.

Parameter Estimates of Regression Paths in the Model of Externalizing Symptoms.

Dependent Variable	Parameter Estimates
Predictor	Coefficient (SE)	β	95% CI
Dissociative symptoms
Child sex	–.14 (.35)	–.01	[–.72, .41]
Income-to-needs ratio	–.56* (.17)	–.11	[–.83, –.30]
Race	1.05* (.40)	.09	[.39, 1.71]
Maternal education (below high school)	1.42* (.50)	.10	[.58, 2.27]
Maternal education (college or above)	–1.62* (.38)	–.10	[–2.22, –1.03]
Total parental aggression (path a)	1.28* (.34)	.17	[.77, 1.84]
Externalizing symptoms in PK
Child sex	.28* (.04)	.26	[.21, .34]
Income-to-needs ratio	–.03 (.02)	–.06	[–.05, .01]
Race	.13* (.04)	.12	[.05, .19]
Maternal education (below high school)	.05 (.05)	.04	[–.03, .14]
Maternal education (college or above)	–.15* (.05)	–.10	[–.24, –.07]
Total parental aggression	.05* (.03)	.08	[.01, .10]
Externalizing symptoms in school years
Child sex	.19* (.03)	.20	[.14, .25]
Income-to-needs ratio	–.03* (.02)	–.09	[–.06, –.01]
Race	.08* (.03)	.08	[.03, .14]
Maternal education (below high school)	.06 (.03)	.05	[–.005, .11]
Maternal education (college or above)	–.16* (.05)	–.12	[–.24, –.09]
Externalizing symptoms in pre-K	.40* (.04)	.44	[.32, .47]
Dissociative symptoms (path b2)	.01* (.003)	.08	[.002, .01]
Total parental aggression (path c2)	.05* (.02)	.08	[.02, .08]

Note. *p < .05.

Coefficient = Unstandardized coefficient; SE = Standard error; β = Standardized coefficient; CI = Bootstrapped confidence interval.

Child sex was coded as 0 (female) and 1 (male). Race was dichotomously coded as 0 (Caucasian) and 1 (African American and others).

In the model of internalizing symptoms (refer to Table 2 and Figure 2a), parental aggression predicted more dissociative symptoms before school entry (path a) but was not associated with internalizing symptoms at any time point. Dissociative symptoms did not correlate with concurrent internalizing symptoms, nor did it predict later symptoms in school years (path b1). Thus, parental aggression did not have significant indirect effect (through dissociative symptoms; a*b1, unstandardized coefficient = .002, SE = .002, bootstrapped p-value = .27, bootstrapped 95% confidence interval = [–.002, .006]) or direct effect (path c1) on internalizing symptoms. No significant sex or race differences were found in internalizing symptoms. Lower income-to-needs ratio and primary caregivers not having a college degree were not related to the level of symptoms prior to school entry but did predict more internalizing symptoms in school years.

The model of externalizing symptoms was displayed in Table 3 and Figure 2b. Higher levels of parental aggression predicted more dissociative symptoms before school entry (path a), which were further related to more externalizing symptoms in school years (path b2). The indirect effect (a*b2) was statistically significant (unstandardized coefficient = .008, SE = .004, bootstrapped p-value = .01, bootstrapped 95% confidence interval = [.002, .02]), and there was a remaining direct effect of parental aggression on externalizing symptoms (path c2). These associations were found after controlling for the relation between parental aggression and externalizing symptoms in pre-K, the correlation between dissociative and externalizing symptoms in pre-K, and the auto-regression between earlier and later externalizing symptoms. Thus, dissociative symptoms before school entry partially mediated the relation between parental aggression in early life and externalizing symptoms in school years, above and beyond the predictive value of externalizing symptoms before school entry. Boys and minority children showed more externalizing symptoms before school entry and in school years, while primary caregiver holding a college degree significantly lowered the risk. Income-to-needs ratio was not related to symptoms in pre-K but was associated with more externalizing symptoms in school years.

Discussion

Exposure to parental aggression in early life presents considerable risk for developing psychopathological symptoms among children and adolescents (Melançon & Gagné, 2011; Riina et al., 2014). To inform prevention practices, it is critical to understand the developmental mechanisms and intermediate phenotypes that lead to increased risk for later symptoms. This study focused on dissociative symptoms among young children that may develop in response to parental aggression and examined whether they mediated risk for internalizing and externalizing symptoms in school years. Using prospectively collected data from a large sample of rural families, we found evidence for mediational relations with externalizing, but not internalizing symptoms. Dissociative symptoms in preschool predicted persistent externalizing symptoms over time as children attended grade school. These findings extend previous research that predominantly used retrospective methods to study trauma-related dissociation and distal outcomes in adolescents and adults. The current study has implications for the prevention of psychological maladjustment in young children who are exposed to parental aggression and calls for further attention to the specificity of mechanisms in clinical and community samples.

Parental Aggression, Dissociation, and Psychopathological Symptoms

Consistent with hypothesis, we found that children who were exposed to higher levels of parental aggression in the first three years of life showed more dissociative symptoms and externalizing symptoms before school entry. Interestingly, both dissociative symptoms and externalizing symptoms measured before school entry accounted for unique variances in externalizing symptoms in later school years. That is, after controlling for the levels of externalizing symptoms children were already displaying before starting school, those who showed early dissociative symptoms were at greater risk for persistent or escalated impulsivity and behavioral problems in school. Although early symptoms are often the best predictors for later symptoms, there is considerable variability in children’s developmental trajectories. Longitudinal evidence has indicated that about half of children who present with externalizing symptoms in early childhood show improvement in behaviors over time, while others’ symptoms persist or escalate (Fanti & Henrich, 2010). Our findings suggest that exposure to parental aggression in early life increases the risk for sustained symptom severity as children transition into grade school, and dissociative symptoms may be a mediating mechanism and early indicator of such risk. The effect size of the association between early dissociative symptoms and later externalizing symptoms was small, warranting caution in the interpretation of its clinical significance. However, it should be noted that we took a conservative approach by including externalizing symptoms before school entry and allowing it to covary with dissociative symptoms. Thus, despite the small effect size, findings suggest that early dissociative symptoms signal underlying risks that explain unique variance in the developmental progression of externalizing symptoms.

Previous research has discussed the overlaps in the behavioral manifestations and high rates of comorbidity between trauma symptoms and externalizing symptoms (e.g., Ford et al., 2000; Szymanski et al., 2011). Our findings further stress the need to study the underlying, and potentially heterogenous mechanisms of children’s externalizing symptoms, rather than just focusing on presenting symptoms (Cuthbert, 2014). In particular, in addition to identifying children with early-onset hyperactivity, impulsivity, aggression, and opposition, programs aiming to prevent externalizing symptoms in school years may also screen for exposure to parental aggression and early dissociative symptoms. Services for children exposed to parental aggression could target generalized dissociative behaviors, and promote children’s use of adaptive coping strategies (e.g., use of adaptive emotion regulation strategies) when facing stressors. It is also important to facilitate the development of abilities that may be influenced by early dissociative behaviors, including empathetic responses and regulation of behaviors.

Surprisingly, we did not find significant associations between parental aggression or dissociative symptoms and children’s internalizing symptoms at any time point. This contradicts findings in some previous studies, which suggested that early violence exposure and dissociation were correlated with more internalizing symptoms (Ensink et al., 2017; Hagan et al., 2015). However, these studies might be subject to reporting bias by collecting information on dissociative and internalizing symptoms from a single respondent (i.e., either mother or teacher); they also failed to control for demographic confounders (e.g., poverty). Studies with traumatized samples did suggest that dissociation plays a role in the development of post-traumatic stress disorders (e.g., Briere et al., 2005), which were not fully captured in our measure of more general anxious and depressive symptoms. Furthermore, young children’s emotional symptoms may be less visible to teachers, and may often manifest as hypervigilance, tantrums, and/or conflicts with peers, which may only be identified by caregivers and teachers as externalizing symptoms. Future studies could look into indices that integrate multiple respondents’ reports (e.g., from parents, peers, and children themselves) and the underlying sources of acting-out behaviors.

Understanding Early Disparities in Psychosocial Adjustment

This study also highlights the necessity to study children’s adjustment in early school years, especially to understand the environmental adversity in early life that impacts this critical period of transition. As evident in Figure 1, the majority of children in this sample experienced some physical and/or psychological aggression from parents, which is consistent with findings of other prevalence surveys (e.g., Runyan et al., 2010). Furthermore, there was a clustering effect of risks, such that children who experienced more frequent parental aggression were also exposed to more types of aggressive behaviors. Additionally, parental aggression often coexisted with other demographic factors associated with disparities in psychosocial adjustment (e.g., poverty, racial minority; refer to Appendix B). These findings call for further research and preventive efforts on the accumulative effect of risk factors.

Family income-to-needs ratio was not related to children’s psychopathological symptoms before school entry but was associated with internalizing and externalizing symptoms in school years. Lower income-to-needs ratio, which may be linked to exposure to various stressors and traumatic experiences (e.g., violence in the neighborhood), was also related to more dissociative symptoms. This sheds light on how poverty in early life is related to growing disparity over time and that the transition to school is a critical window for prevention, as it presents extra challenge for children from socioeconomically disadvantaged families. Furthermore, African American children were at higher risk for dissociative and externalizing symptoms, even when the effects of income-to-needs ratio and parental aggression were statistically accounted for. Previous studies indicated that African American families in the current sample faced higher levels of neighborhood disadvantage (Vernon-Feagans et al., 2013) and had less access to stable and high-quality child care (Bratsch-Hines et al., 2015; Vernon-Feagans et al., 2013). These factors may have contributed to the gaps in children’s adjustment in school, and comprehensive efforts are needed to address the racial disparities in families’ access to child care and social services.

It is also worth noting that, although no sex difference was found for dissociative symptoms, boys showed more externalizing symptoms than girls before school entry, and the gap further widened in school years. The literature has consistently shown higher levels of early-onset externalizing symptoms among boys in comparison to girls, especially based on teachers’ reports (Achenbach et al., 2008). However, evidence is mixed regarding the developmental changes in externalizing symptoms across the transition to school, with some studies showing that boys are at higher risk for elevating externalizing symptoms (e.g., Zhou et al., 2007), whereas others showing the gap between boys and girls closing over time (e.g., Godinet et al., 2014). Further studies are needed to examine whether the association between sex and developmental variations in externalizing symptoms vary systematically by socialization contexts or other characteristics of the samples.

Limitations and Implications

The current study has some limitations that warrant caution in the interpretation and generalization of findings. First, our measure of parental aggression did not separate different dimensions of exposure (e.g., severity, frequency), which may have differential effects on child outcomes. Second, this measure relied on primary caregivers’ report, which was efficient and had the potential to characterize the actual amount of aggressive behaviors (compared to reports from child protective agencies), but might be subject to social desirability bias. Finally, the current study did not obtain measures of parents’ dissociative symptoms, which would be important to account for in future studies. Despite the limitations, this study provides one of the first pieces of prospective longitudinal evidence on how dissociative symptoms mediate the effects of parental aggression in early life on children’s externalizing symptoms during school years. These findings call for attention on how early life experiences contribute to disparities in school adjustment and highlight the importance of taking a trauma-informed approach when developing preventive services to reduce children’s risks for attentional and behavioral problems.

Footnotes

Acknowledgments

We thank the families and research assistants who made this study possible.

Authors’ Note

This study is part of the Family Life Project.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Funding for this study was provided by grants from the National Institute of Child Health and Human Development P01HD039667; HD080786.

ORCID iD

Xutong Zhang

Author Biographies

Xutong Zhang, MS, is a doctoral candidate at the Department of Human Development and Family Studies, The Pennsylvania State University. Her research focuses on children’s emotional development in the context of parent-child interaction, as well as the roles of stress and parental self-regulation in shaping parenting behaviors and risk for abuse and neglect.

Lisa M. Gatzke-Kopp, PhD, is a professor of Human Development and Family Studies at The Pennsylvania State University. Her research focuses on the environmental and experiential factors that shape children’s development and risk for psychopathology.

Appendix A. Distribution of Child Dissociative Symptoms in the Current Sample

Table B1. Parameter Estimates of Correlations and Variances in the Model of Internalizing Symptoms.

Correlation Path	Parameter Estimates
Correlation Path	Coefficient	Standard Error	p
Dissociative symptoms–Internalizing symptoms in PK	.07	.04	.10
Total parental aggression–Child sex	.08	.03	<.01
Total parental aggression–Race	.14	.03	<.01
Total parental aggression–Income-to-needs ratio	–.06	.03	.02
Total parental aggression–M education (below high school)	–.02	.03	.50
Total parental aggression–M education (college or above)	–.04	.02	.10
Income-to-needs ratio–Race	–.41	.02	<.01
Income-to-needs ratio–M education (below high school)	–.36	.02	<.01
Income-to-needs ratio–M education (college or above)	.58	.03	<.01
Race–M education (below high school)	.13	.03	<.01
Race–M education (college or above)	–.26	.02	<.01
M education (below high school)–M education (college or above)	–.23	.01	<.01
Variable	Variance	Standard Error	p
Total parental aggression	.63	.07	<.01
Dissociative symptoms	31.55	2.54	<.01
Internalizing symptoms in PK	.11	.01	<.01
Internalizing symptoms in school years	.10	.01	<.01
Child sex	.25	<.001	<.01
Race	.25	.002	<.01
Income-to-needs ratio	1.47	.07	<.01
M education (below high school)	.18	.01	<.01
M education (college or above)	.12	.01	<.01

Notes. PK = Pre-kindergarten; M education = Maternal education.

Figure A1. Distribution of child dissociative symptoms.

Note. The x-axis represents the total score on the Child Dissociative Checklist (possible range = 0–40). In the current sample, 12.5% of the children scored within the clinical range (≥ 12).

Appendix B. Correlation Paths and Variances of Variables in the SEM Models.

Table B2. Parameter Estimates of Correlations and Variances in the Model of Externalizing Symptoms.

Correlation Path	Parameter Estimates
Correlation Path	Coefficient	Standard Error	p
Hyperactivity in PK–Hyperactivity in school years	.23	.14	.09
Conduct problems in PK–Conduct problems in school years	.32	.06	<.01
Dissociative symptoms–Externalizing symptoms in PK	.19	.04	<.01
Total parental aggression–Child sex	.08	.03	<.01
Total parental aggression–Race	.14	.03	<.01
Total parental aggression–Income-to-needs ratio	–.06	.03	.02
Total parental aggression–M education (below high school)	–.02	.03	.50
Total parental aggression–M education (college or above)	–.04	.02	.10
Income-to-needs ratio–Race	–.41	.02	<.01
Income-to-needs ratio–M education (below high school)	–.36	.02	<.01
Income-to-needs ratio–M education (college or above)	.58	.03	<.01
Race–M education (below high school)	.13	.03	<.01
Race–M education (college or above)	–.26	.02	<.01
M education (below high school)–M education (college or above)	–.23	.01	<.01
Variable	Variance	Standard Error	p
Total parental aggression	.63	.07	<.01
Dissociative symptoms	31.56	2.54	<.01
Externalizing symptoms in PK	.25	.03	<.01
Externalizing symptoms in school years	.13	.01	<.01
Hyperactivity in PK	.05	.03	.05
Hyperactivity in school years	.07	.02	<.01
Conduct problems in PK	.07	.01	<.01
Conduct problems in school years	.06	.01	<.01
Child sex	.25	<.001	<.01
Race	.25	.002	<.01
Income-to-needs ratio	1.47	.07	<.01
M education (below high school)	.18	.01	<.01
M education (college or above)	.12	.01	<.01

Note. PK = Pre-kindergarten; M education = Maternal education.

References

Achenbach

T. M.

, Becker

, Döpfner

, Heiervang

, Roessner

, Steinhausen

H. C.

, & Rothenberger

(2008). Multicultural assessment of child and adolescent psychopathology with ASEBA and SDQ instruments: Research findings, applications, and future directions. Journal of Child Psychology and Psychiatry , 49, 251–275.

American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (5th ed.). American Psychiatric Publishing.

Atkins

M. S.

, Cappella

, Shernoff

E. S.

, Mehta

T. G.

, & Gustafson

E. L.

(2017). Schooling and children’s mental health: Realigning resources to reduce disparities and advance public health. Annual Review of Clinical Psychology , 13, 123–147.

Bandura

(1978). Social learning theory of aggression. Journal of Communication , 28, 12–29.

Bentler

P. M.

(2007). On tests and indices for evaluating structural models. Personality and Individual Differences, 42, 825–829.

Bratsch-Hines

M. E.

, Mokrova

, & Vernon-Feagans

L.,

FLP Key Investigators (2015). Child care instability from 6 to 36 months and the social adjustment of children in prekindergarten. Early Childhood Research Quarterly , 30, 106–116.

Briere

, Scott

, & Weathers

(2005). Peritraumatic and persistent dissociation in the presumed etiology of PTSD. American Journal of Psychiatry , 162, 2295–2301.

Calkins

S. D.

(2007). The emergence of self-regulation: Biological and behavioral control mechanisms supporting toddler competencies. In Brownell

C. A.

& Kopp

C. B.

(Eds.), Socioemotional development in the toddler years: Transitions and transformations (pp. 261–284). Guilford Press.

Centers for Disease Control and Prevention. (n.d.). Data and statistics on children’s mental health. Retrieved, July 12, 2020, from https://www.cdc.gov/childrensmentalhealth/data.html

10.

Cuthbert

B. N.

(2014). The RDoC framework: Facilitating transition from ICD/DSM to dimensional approaches that integrate neuroscience and psychopathology. World Psychiatry , 13, 28–35.

11.

Dalenberg

C. J.

, Brand

B. L.

, Gleaves

D. H.

, Dorahy

M. J.

, Loewenstein

R. J.

, Cardena

, Frewen

P. A.

, Carlson

E. B.

, & Spiegel

(2012). Evaluation of the evidence for the trauma and fantasy models of dissociation. Psychological Bulletin , 138, 550–588.

12.

Danese

, & McEwen

B. S.

(2012). Adverse childhood experiences, allostasis, allostatic load, and age-related disease. Physiology & Behavior , 106, 29–39.

13.

Denham

S. A.

, Bassett

H. H.

, Zinsser

, & Wyatt

T. M.

(2014). How preschoolers’ social-emotional learning predicts their early school success: Developing theory-promoting, competency-based assessments. Infant and Child Development , 23, 426–454.

14.

Enders

C. K.

(2010). Applied missing data analysis . Guilford Press.

15.

Ensink

, Berthelot

, Bégin

, Maheux

, & Normandin

(2017). Dissociation mediates the relationship between sexual abuse and child psychological difficulties. Child Abuse & Neglect , 69, 116–124.

16.

Falk

C. F.

(2018). Are robust standard errors the best approach for interval estimation with nonnormal data in structural equation modeling? Structural Equation Modeling: A Multidisciplinary Journal , 25, 244–266.

17.

Fan

, Thompson

, & Wang

(1999). Effects of sample size, estimation methods, and model specification on structural equation modeling fit indexes. Structural Equation Modeling: A Multidisciplinary Journal , 6, 56–83.

18.

Fanti

K. A.

, & Henrich

C. C.

(2010). Trajectories of pure and co-occurring internalizing and externalizing problems from age 2 to age 12: Findings from the National Institute of Child Health and Human Development Study of Early Child Care. Developmental Psychology , 46, 1159–1175.

19.

Ford

J. D.

, Racusin

, Ellis

C. G.

, Daviss

W. B.

, Reiser

, Fleischer

, & Thomas

(2000). Child maltreatment, other trauma exposure, and posttraumatic symptomatology among children with oppositional defiant and attention deficit hyperactivity disorders. Child Maltreatment , 5, 205–217.

20.

Gershoff

E. T.

(2002). Corporal punishment by parents and associated child behaviors and experiences: A meta-analytic and theoretical review. Psychological Bulletin , 128, 539–579.

21.

Godinet

M. T.

, Li

, & Berg

(2014). Early childhood maltreatment and trajectories of behavioral problems: Exploring gender and racial differences. Child Abuse & Neglect , 38, 544–556.

22.

Goodman

(1997). The strengths and difficulties questionnaire: A research note. Journal of Child Psychology and Psychiatry , 38, 581–586.

23.

Hagan

M. J.

, Hulette

A. C.

, & Lieberman

A. F.

(2015). Symptoms of dissociation in a high-risk sample of young children exposed to interpersonal trauma: Prevalence, correlates, and contributors. Journal of Traumatic Stress , 28, 258–261.

24.

Hulette

A. C.

, Freyd

J. J.

, & Fisher

P. A.

(2011). Dissociation in middle childhood among foster children with early maltreatment experiences. Child Abuse & Neglect , 35, 123–126.

25.

Jonson-Reid

, Kim

, Barolak

, Citerman

, Laudel

, Essma

, Fezzi

, Green

, Kontak

, Mueller

, & Thomas

(2007). Maltreated children in schools: The interface of school social work and child welfare. Children & Schools , 29, 182–191.

26.

Lai

(2018). Estimating standardized SEM parameters given nonnormal data and incorrect model: Methods and comparison. Structural Equation Modeling: A Multidisciplinary Journal , 25, 600–620.

27.

Lee

S. J.

, Taylor

C. A.

, Altschul

, & Rice

J. C.

(2013). Parental spanking and subsequent risk for child aggression in father-involved families of young children. Children and Youth Services Review , 35, 1476–1485.

28.

Linning

L. M.

, & Kearney

C. A.

(2004). Post-traumatic stress disorder in maltreated youth: A study of diagnostic comorbidity and child factors. Journal of Interpersonal Violence , 19, 1087–1101.

29.

Lyons-Ruth

, Dutra

, Schuder

M. R.

, & Bianchi

(2006). From infant attachment disorganization to adult dissociation: Relational adaptations or traumatic experiences? Psychiatric Clinics , 29, 63–86.

30.

Melançon

, & Gagné

M. H.

(2011). Father’s and mother’s psychological violence and adolescent behavioral adjustment. Journal of Interpersonal Violence , 26, 991–1011.

31.

Pears

K. C.

, Fisher

P. A.

, Bruce

, Kim

H. K.

, & Yoerger

(2010). Early elementary school adjustment of maltreated children in foster care: The roles of inhibitory control and caregiver involvement. Child Development , 81, 1550–1564.

32.

Putnam

F. W.

(2006). Dissociative disorders. In Cicchetti

& Cohen

D. J.

(Eds.), Developmental Psychopathology (Vol. 3, 2nd ed., pp. 657–695). Wiley.

33.

Putnam

F. W.

, Helmers

, & Trickett

P. K.

(1993). Development, reliability, and validity of a child dissociation scale. Child Abuse & Neglect , 17, 731–741.

34.

Putnam

F. W.

, & Peterson

(1994). Further validation of the Child Dissociative Checklist. Dissociation: Progress in the Dissociative Disorders , 7, 204–211.

35.

Riina

E. M.

, Martin

, & Brooks-Gunn

(2014). Parent-to-child physical aggression, neighborhood cohesion, and development of children’s internalizing and externalizing. Journal of Applied Developmental Psychology , 35, 468–477.

36.

Rosseel

(2012). Lavaan: An R package for structural equation modeling and more. Journal of Statistical Software , 48, 1–36.

37.

Runyan

D. K.

, Shankar

, Hassan

, Hunter

W. M.

, Jain

, Paula

C. S.

, Bangdiwala

S. I.

, Ramiro

L. S.

, Muñoz

S. R.

, Vizcarra

, & Bordin

I. A.

(2010). International variations in harsh child discipline. Pediatrics , 126, 701–711.

38.

Savalei

(2010). Expected versus observed information in SEM with incomplete normal and nonnormal data. Psychological Methods , 15, 352–367.

39.

Sedlak

A. J.

, Mettenburg

, Basena

, Petta

, McPherson

, Greene

, & Li

(2010). Fourth National Incidence Study of Child Abuse and Neglect (NIS–4): Report to Congress. Department of Health and Human Services, Administration for Children and Families .

40.

Silberg

J. L.

(2014). Dissociative disorders in children and adolescents. In Lewis

& Rudolph

K. D.

(Eds.), Handbook of developmental psychopathology (pp. 761–775). Springer.

41.

Silvern

, & Griese

(2012). Multiple types of child maltreatment, posttraumatic stress, dissociative symptoms, and reactive aggression among adolescent criminal offenders. Journal of Child & Adolescent Trauma , 5, 88–101.

42.

Slep

A. M. S.

, & O’leary

S. G.

(2005). Parent and partner violence in families with young children: Rates, patterns, and connections. Journal of Consulting and Clinical Psychology , 73, 435–444.

43.

Stegge

, & Terwogt

M. M.

(2007). Awareness and regulation of emotion in typical and atypical development. In Gross

J. J

(Ed.), Handbook of emotion regulation (pp. 249–268). Guilford Press.

44.

Straus

M. A.

, Hamby

S. L.

, Finkelhor

, Moore

D. W.

, & Runyan

(1998). Identification of child maltreatment with the Parent–Child Conflict Tactics Scales: Development and psychometric data for a national sample of American parents. Child Abuse & Neglect , 22, 249–270.

45.

Szymanski

, Sapanski

, & Conway

(2011). Trauma and ADHD–Association or diagnostic confusion? A clinical perspective. Journal of Infant, Child, and Adolescent Psychotherapy , 10, 51–59.

46.

Vernon-Feagans

, & Bratsch-Hines

M. E.

, FLP Key Investigators (2013). Caregiver–child verbal interactions in child care: A buffer against poor language outcomes when maternal language input is less. Early Childhood Research Quarterly , 28, 858–873.

47.

Vernon-Feagans

, & Cox

, FLP Key Investigators (2013). The Family Life Project: An epidemiological and developmental study of young children living in poor rural communities. Monographs of the Society for Research in Child Development, 78 , 1–150.

48.

Vonderlin

, Kleindienst

, Alpers

G. W.

, Bohus

, Lyssenko

, & Schmahl

(2018). Dissociation in victims of childhood abuse or neglect: A meta-analytic review. Psychological Medicine , 48, 2467–2476.

49.

Wadsworth

M. E.

(2015). Development of maladaptive coping: A functional adaptation to chronic, uncontrollable stress. Child Development Perspectives , 9, 96–100.

50.

Zhou

, Hofer

, Eisenberg

, Reiser

, Spinrad

T. L.

, & Fabes

R. A.

(2007). The developmental trajectories of attention focusing, attentional and behavioral persistence, and externalizing problems during school-age years. Developmental Psychology , 43, 369–385.