Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2019 Oct 2.
Published in final edited form as: J Trauma Dissociation. 2018 May-Jun;19(3):307–324. doi: 10.1080/15299732.2018.1441353

Polyvictimization and externalizing symptoms in foster care children: The moderating role of executive function

Sarah R Horn 1, Leslie E Roos 1, Kathryn G Beauchamp 1, Jessica E Flannery 1, Philip A Fisher 1
PMCID: PMC6774668  NIHMSID: NIHMS1043796  PMID: 29547079

Abstract

Prior research has identified the role of childhood maltreatment in externalizing problems and executive function (EF) deficits, but minimal work has been done to characterize the effects of co-occurring maltreatment types, defined as polyvictimization. Here, we sought to characterize the association between polyvictimization and externalizing problems in a sample of foster care children aged 3–4 years (N = 84) and examine how EF may mediate or moderate that relationship. A moderation model was supported in that only polyvictimized children with EF scores 1.62 or more standard deviations below the mean were at heightened risk for clinically severe externalizing problems, while no association between polyvictimization and externalizing problems were observed for children who scored at the mean or above on the EF measure. Findings highlight that EF may serve as a resilience factor indicating that individual differences in polyvictimized children’s EF skills help to predict variability in externalizing problems. Future research on designing and optimizing intervention programs that target EF skills may mitigate the development of maladaptive outcomes for polyvictimized children.

Keywords: Polyvictimization, childhood maltreatment, externalizing problems, attention, executive functioning

Young children in foster care are disproportionately more likely to experience polyvictimization (i.e., multiple types of maltreatment; Pears, Kim, & Fisher, 2008) and, in turn, are at elevated risk for negative outcomes (Leeb, Lewis, & Zolotor, 2011). In a cross-sectional survey of children’s exposure to violence, 49% of the children surveyed suffered two or more types of polyvictimization (Finkelhor et al., 2011); for children in foster care, rates of polyvictimization may be as high as 95% (Lau et al., 2005; Manly, Kim, Rogosch, & Cicchetti, 2001). Prior maltreatment research has largely focused on specific forms of abuse or neglect. In contrast, children’s actual experiences often involve more than one type of maltreatment (Greeson et al., 2011; Turner, Finkelhor, & Ormrod, 2010). It is important to note that a wide range of variability exists not only in children’s experiences of victimization types but also in the risk for associated negative outcomes.

A related issue among foster children is the documented presence of externalizing problems (e.g., defiance, aggression, impulsivity) commonly observed above and beyond other mental health issues (Stein, Mazumdar, & Rae-Grant, 1996). Of concern, externalizing symptoms are often precursors to negative outcomes, such as risk-taking, antisocial behavior, and suicidality (Campbell, Shaw, & Gilliom, 2000). Despite numerous studies documenting the high prevalence of externalizing problems in young children in foster care, ranging from 20% to 78% (Keil & Price, 2006), there is relatively little research investigating how variation in children’s maltreatment experiences is associated with externalizing risk within foster care samples. Heterogeneity in individuals’ responses to maltreatment is an important challenge for both researchers and practitioners (Afifi & MacMillan, 2011). Polyvictimization quantifies variability in maltreatment experiences to disentangle individual differences in externalizing problem risk. Focusing on the number of maltreatment types experienced is a promising approach to operationalizing polyvictimization (Cecil, Viding, Fearon, Glaser, & McCrory, 2017).

Polyvictimization and externalizing behavior

Externalizing problems are often observed in samples of polyvictimized children in foster care and community populations (Cyr et al., 2012; Turner et al., 2010). Theoretical perspectives lend insight into this association. From a social learning perspective, externalizing behavior can be conceptualized as modeling of observed parental behavior (Patterson & Yoerger, 2002). Per social learning theory, a child both witnessing domestic violence and experiencing physical abuse may have observed aggression and frequently engaged in coercive parenting interactions that urge the child to imitate the modeled behavior (Moylan et al., 2010). Experiences of multiple maltreatment types increase the repertoire of modeled behaviors from which children learn, thereby increasing the likelihood of similar violence perpetration. Attachment theory suggests that the lack of a responsive attachment figure within contexts of polyvictimization contributes to risk for externalizing behavior via the child’s failure to develop effective self-regulatory emotional and behavioral strategies for dealing with distress, which is likely to be more pervasive across contexts in situations of polyvictimization (Stovall-McClough & Dozier, 2004). While supportive caregiving and authoritative discipline are associated with regulatory capacities (Kochanska, Murray, & Harlan, 2000; Spinrad et al., 2007), harsh discipline is linked to impaired self-regulatory capacities in children, including deficits in inhibitory control and behavioral regulation (Olson et al., 2011). In turn, diminished self-regulatory capacities may increase the risk for externalizing behavior (Fay-Stammbach, Hawes, & Meredith, 2014).

Numerous studies have demonstrated a robust link between early adversity and a range of externalizing behavior problems (e.g., aggression, delinquency). Several mediators and moderators underlying this relationship have been identified, such as emotion regulation (Kim & Cicchetti, 2010), parenting styles (Lansford et al., 2006), and compromised attention (Roman, Ensor, & Hughes, 2016). A study of community children aged 4–15 years found that externalizing problems significantly increased for each additional adverse event exposure (Fleckman, Drury, Taylor, & Theall, 2016). Few studies have investigated the risk of externalizing behavior in the context of polyvictimization; it is essential to further delineate the complex relationship between co-occurring victimizations and externalizing symptomatology.

Executive function as a mediator or moderator in the path from polyvicimtization to externalizing problems

Executive functions (EF), or the higher-order cognitive processes supporting goal-directed behavior, are considered a core capacity predictive of adaptive functioning (Garon, Bryson, & Smith, 2008). EF is considered to comprise at least three key subdomains—inhibitory control, working memory, and mental flexibility (Garon, Bryson, & Smith, 2008). Germane to the present investigation, deficits in EF appear to be strongly related to externalizing behavior in young maltreated children; a recent meta-analysis of primarily community-based studies reported a medium effect size for the association between EF and externalizing behavior (Schoemaker, Mulder, Deković, & Matthys, 2013). Attention is fundamentally involved in several EF component operations (Barkley, 1996) and the development of EF skills in pre-school-aged children may be in large part due to the role of attention (Garon et al., 2008). Further, performance on EF tasks is highly correlated with attention processes (Kane & Engle, 2003; Miller & Marcovitch, 2015).

The extent to which EF may be a mediator or a moderator in the association between polyvictimization and externalizing problems is a critical open question. Most extant work on this topic has focused on EF’s role in the association between parental risk and externalizing behavior, which may inform the development of conceptual models to describe how a range of early experiences impact EF and, in turn, externalizing behavior. Child EF at age 3 has been shown to mediate the relationship between maternal depressive symptoms and externalizing problems at age 6 years (Roman et al., 2016). In a sample of families living in rural poverty, EF mediated the association between sensitive parenting and lower externalizing behavior longitudinally across early childhood (Sulik, Blair, Mills-Koonce, Berry, & Greenberg, 2015). These data suggest that EF skills may explain the link between maltreatment and externalizing behavior.

Evidence also exists for EF as a moderator, such that children with high EF demonstrate resilience following maltreatment, including lower risk for externalizing behavior, compared to children with low EF (Obradović, 2010). Attention skills have also been shown to moderate the link between maltreatment intensity and academic performance (Slade & Wissow, 2007). Such findings provide support for a resilience factor model, in which certain factors can either buffer children or heighten the risks of the negative impacts of early life adversity (Masten, 2001). EF skills are hypothesized to serve as a protective or risk factor underlying the link between polyvictimization and externalizing behaviors, in that deficits in EF place young children at particularly high risk for maladaptive outcomes while above average EF skills may buffer against the negative impacts of early life adversity.

Current study

Given the limited previous research examining links between polyvictimization and externalizing symptoms within maltreated samples, we first sought to examine the extent to which polyvictimization increased risk for externalizing problems (defined as those in the borderline, or clinical range) in a secondary data analysis from a sample of foster care children. Polyvictimization was operationalized as the number of different maltreatment types experienced. We hypothesized that children with higher polyvictimization would be at elevated risk for externalizing problems. Next, we tested the relationship between polyvictimization and externalizing behaviors and investigated the role of EF in mediation and moderation models. Given support for both models in the literature, and a dearth of studies in polyvictimized samples, we did not have a specific hypothesis regarding whether EF would mediate or moderate the association between polyvictimization and externalizing problems. A mediation model would suggest that EF accounts for the link between polyvictimization and externalizing behaviors while a moderation model would indicate individual differences in children’s EF skills place them at variable risk for externalizing problems, given a history of polyvictimization (see Figure 1).

Figure 1.

Figure 1

Open in a new tab

Hypothesized mediation and moderation models.

Method

Participants

The sample comprised a subgroup of participants from a randomized controlled trial (RCT) of Multidimensional Foster Care for Preschoolers (MTFC-P; Fisher, Gunnar, Chamberlain, & Reid, 2000), which included children aged 3–6 years who were entering a new foster care placement between 2000 and 2003. The original sample included 117 maltreated foster care children (63 males) recruited from a mid-sized urban area in the Pacific Northwest participating in the RCT. For this study, we combined foster treatment and foster comparison arms into a single group for analysis. An additional 60 children were recruited from the community as a comparison sample. The community sample was recruited from the same geographic area and from low income households. The “community controls” had no history of child welfare involvement. While community controls are not the focus of the present study as they had no polyvictimization data, they are briefly included throughout the methods and measures to demonstrate the extent to which children in foster care exhibited functional impairment compared to community controls.

The present study analyzed foster care and community children aged 3–4 years due to an age restriction on the EF measure. Therefore, the sample was reduced to N = 88 foster care children and € = 47 community controls. For analyses comparing foster care children to the community controls on externalizing problems and EF scores, the full subsample was utilized. However, for analyses examining polyvictimization, EF, and externalizing problems, we only used the foster care sample.

Sociodemographic variables

The ethnicity of the sample reflects the local community: In the foster care sample, the ethnicity breakdown was: 88.9% Caucasian, 4.3% Native American, 5.1% mixed race/other, .9% Latino, and .9% African-American. In the community sample, 76.7% of the children were Caucasian, 6.7% Native-American, 6.7% mixed race/other, 1.7% Latino, and 6.7% African-American. In the foster care sample, 53.8% (n = 63) of the children were male and the average age was 4.04 years (SD = .85). In the community sample, the average age was 4.33 (SD = 0.79) and 53.5% of the sample was male (n = 32).

Measures

Study measures were collected from children and foster parents in laboratory and home-based assessments after obtaining informed consent to participate in the research study. All procedures were developed in partnership with the local and state child welfare agencies and were approved by the Institutional Review Board.

Maltreatment profiles

The Maltreatment Classification System (MCS; Barnett, Manly, & Cicchetti, 1993) was utilized to code child welfare system (CWS) case records for lifetime maltreatment histories. The MCS codes for several types of maltreatment, including: physical abuse, sexual abuse, failure to provide (e.g., parental failure to provide adequate food, clothing, or a safe living environment), supervisory neglect (e.g., parental failure to provide age-appropriate supervision), emotional maltreatment (e.g., parental rejection, allowing the child to witness traumatic events), educational maltreatment (i.e., parental failure to send the child to school), and moral/legal maltreatment (i.e., parents using the child for illegal purposes). The rates for the latter two categories were extremely low: educational maltreatment (n = 9) and moral/legal maltreatment (n = 1). Because of the low base rates and the distinct nature of these categories, these two maltreatment types were dropped from further score calculation and analyses. The most common types of maltreatment experienced for the children aged 3–4 years were emotional neglect (90.9%), supervisory neglect (88.6%), and physical neglect (79.5%). Roughly one quarter of the foster care sample experienced physical or sexual abuse (Table 1).

Table 1.

Descriptive statistics for the maltreatment variables.

Maltreatment type Number of children Percent of incidence
Physical abuse 24 27.3
Sexual abuse 20 22.7
Physical neglect 70 79.5
Supervisory neglect 78 88.6
Emotional maltreatment 80 90.9
Polyvictimization rates
One maltreatment type 4 4.5
Two maltreatment types 20 22.7
Three maltreatment types 34 38.6
Four maltreatment types 24 27.3
Five maltreatment types 6 6.8
Open in a new tab

Coding process

Caseworkers and the local child welfare agency consented to the review of case records. A child welfare representative excluded identifiable information. Coders completed extensive training on confidentiality agreements. Narratives of CWS referrals contained material about the specific types of maltreatment reported and investigated; however, narratives did not necessarily contain information on each type of maltreatment coded per MCS. Coders identified distinct maltreatment incidents from the narratives. To qualify as an incident, the event had to fit the MCS definitions of maltreatment and had to be reported by a mandatory reporter or caseworker. Maltreatment incidents were coded per type using the MCS.

Case records were coded by two coders who had been trained by an MCS author (Manley). To compute interrater agreement, 20% of the case records were double-coded. Overall, agreement on identification of incidents was high (80%). The average kappa was .72 across all the categories (physical abuse κ = .82, sexual abuse κ = .67, failure to provide κ = .65, supervisory neglect κ = .65, emotional maltreatment κ = .79).

Polyvictimization

Polyvictimization was conceptualized as a continuous measure of number of types of maltreatment experienced. Subject maltreatment types were coded from 0 to 5, with 0 indicating no maltreatment up to the maximum of 5 discrete maltreatment types. Overall, polyvictimization was frequent, with children experiencing an average of three different types of maltreatment. To note, 95.5% of this sample experienced polyvictimization, which is comparable to other maltreated samples (Lau et al., 2005; Manly et al., 2001).

Executive functioning

NEPSY: a developmental neuropsychological assessment

Children completed the NEPSY (Korkman, Kemp, & Kirk, 1998), which includes a series of 36 subtests assessing neuropsychological development across five domains, from which we specifically examined the Attention/Executive Functioning domain (other domains: Language, Sensorimotor, Visuospatial, and Memory and Learning). The EF domain assesses children’s inhibition, self-regulation, and selective and sustained attention through two subtests.

Visual attention subtest

This subtest assesses children’s speed and accuracy by instructing children to maintain attention and focus selectively on a visual target within a display. The first task is a simple selective attention task. The child is instructed to quickly locate and mark a picture of a bunny from various pictures in a linear array. Additionally, the child must search for pictures of a cat embedded within a random array of varying pictures.

Statue subtest

The statue subtest is designed to assess motor persistence and inhibition. Children are instructed to maintain a body position with their eyes closed for a 75 second period and to restrain any impulse to respond to sound distracters. Sound distractors include a pencil dropping, coughing, knocking on the table twice, saying “ho hum” and “time’s up.” A low score is indicative of poor inhibitory control and motor persistence.

Children in the foster care sample had a significantly lower average EF NEPSY scores of 93.11 (SD = 15.02) compared to the community sample average score of 99.24 (SD = 12.75) (t(128) = −2.34, p = 0.021).

Externalizing problems

Child behavior checklist

The CBCL (Achenbach, 1991) is a 112-item caregiver report for identifying problem behaviors in children. Caregivers rated children’s behavior over the preceding 2 months on a 3-point scale (not true, somewhat/sometimes true, and very true/often true). The items yield a total score and two composite scores – Internalizing and Externalizing. The Externalizing subscale includes symptoms related to hyperactivity, defiance, aggression, and noncompliance. The CBCL externalizing subscale was dichotomized using a cut-off of T-scores ≥60, following the recommended guidelines for referral to clinical services.

In the foster sample, 54.5% (N = 48) of the children met the cut-off indicating clinical levels of externalizing behavior symptomatology compared to 12.8% of the community sample (N = 6; t(133) = 5.13, p < 0.001).

Data analysis plan

We first assessed bivariate correlations between the variables of interest in the foster care children. This step included bivariate correlations with sociodemographic and CWS-related covariates to assess which covariates were related to EF and/or externalizing behaviors. All covariates significantly associated with either variable were included in subsequent moderation and mediation analyses. Next, we tested the relationship between polyvictimization and externalizing behaviors, investigating the role of EF in mediation and moderation models. Specifically, we investigated the extent to which EF served as a mediator of the hypothesized association between polyvictimization and externalizing problems. We further investigated EF as a moderator of the hypothesized association between polyvictimization and externalizing problems, testing if high EF would be a protective factor from the presence of externalizing behaviors given a polyvictimization history.

Results

Preliminary covariate analyses

Zero-order correlations between all variables of interest and potential confounders, including age, number of out-of-home placements, maternal education, and gross annual household income were conducted (Table 2). Gender differences were examined with an independent t-test. Lower gross annual household income was positively correlated with externalizing problems (r = 0.31, p < .001). Age, number of transitions, and maternal education were not significantly correlated with any variables of interest (p-values > .083). There was a significant effect of gender on EF performance with boys performing worse than girls (t(128) = −3.78, p < .001). Accordingly, gender and annual household income were considered as covariates in subsequent multivariate analyses.

Table 2.

Zero-order correlations among the variables.

Variable 1 2 3 4 5 6 7 8
1 Age - .09 −.08 −.01 .09 .12 .16 .15
2 Gender - - .05 .06 −.01 .07 .32* .14
3 Maternal education - - - .14 −.13 .01 .03 .01
4 Household income - - - - .13 .30 −.07 .09
5 Number of transitions - - - - - .09 −.18 .11
6 Externalizing Problems - - - - - - −.13 .16
7 EF NEPSY Score - - - - - - - .09
8 Polyvictimization - - - - - - - -
Open in a new tab
*

p < .001

Multivariate analyses

To test baseline EF as a potential mediator or moderator linking polyvictimization to elevated externalizing problems, the PROCESS add-on module of SPSS v.22 was utilized (Hayes, 2013).

Missing data

Four children did not have complete data on the EF domain and were dropped from multivariate analyses due to listwise deletion procedures. Children who were missing data on the NEPSY test were more likely to have primary caregivers with lower education (t(134) = 59.03, p < 0.001). For all multivariate analyses, a sample size of N = 84 was included.

Mediation analysis

Polyvictimization was not associated with externalizing problems in initial bivariate correlations (R2 = .16, p = .15; Figure 1, Path-C). Further, there was no association between polyvictimization and EF (R2 = .09, p = .43); Figure 1, Path-A) or EF and externalizing (R2 = −13, p = .14; Figure 1, Path-B). The model predicting EF from polyvictimization, gender, and household income was significant (F(3, 76) = 5.07, R2 = .18, p = .003). Male gender predicted externalizing (β = 11.54, SE = 3.09, p < .001); no other predictors were significant. The overall model predicting externalizing from polyvictimization, gender, income, and EF was not significant (Cox & Snell’s R2 = .05, p = .35). No relation was observed between externalizing problems and any other variables, although there was a trend level direct effect for polyvictimization on externalizing in the full model (β = .43, SE = .26, p = .09). No evidence was observed for an indirect effect of polyvictimization on externalizing problems through EF (β = −.03, bootstrap SE = .07, bootstrap CI = [−0.25 to −0.04], indicating no support for a mediation model.

Moderation analysis

A moderation analysis was conducted predicting externalizing problems from polyvictimization, EF, an interaction of polyvicimtization×EF, considering gender and household income covariates. In this model, polyvictimization was a trend level predictor of externalizing problems (β = .49, SE = .28, p = .08). The polyvicimtization×EF interaction was a significant predictor of externalizing problems (β = −.05, SE = .02, p = .026). Neither EF nor the covariates were significant predictors of externalizing problems. Although the full logistic model, with covariates, was close to significant (Cox & Snell’s R2 = 0.12, p = .07), results indicated that this overall model did not explain significant variance in externalizing problems. To improve data fit, we removed non-significant covariates (gender and income; p > .38), resulting in a significant model fit (Cox & Snell’s R2 = .11, p = .02) with a similar pattern observed across predictors. In this final model, there was a trend level effect of polyvictimization on externalizing problems (β = .47, SE = .26, p = .08) and the interaction of polyvicimtization×EF remained significant (β = −.04, SE = .02, p = .04). There was no significant direct effect of EF on externalizing problems (β = −028, SE = .17, p = .10).

Conditional effects probing the polyvicimtization×EF interaction from the final model are shown in Figure 2. The interaction was probed by testing the conditional effects of externalizing behavior at three levels of EF: one standard deviation (SD) below the mean (score = 78.09), at the mean (M = 93.11), and one SD above the mean (score = 108.13). As demonstrated in Table 3, externalizing problems were significantly related to polyvicimtization only when EF was below the mean (β = 1.06, SE = 0.42, p = .01), but not when EF was at or above the mean. The Johnson–Neyman technique demonstrated that the relationship between externalizing problems and polyvictimization was significant when EF scores fell 1.62 SD below the mean score of 93.11 (e.g., an EF score ≤ 91.49), but insignificant with EF scores higher than 91.49.

Figure 2.

Figure 2

Open in a new tab

Conditional effects of Polyvictimization × EF on Externalizing Problems.

Table 3.

Conditional effects of polyvictimization on externalizing problems at different levels of executive function.

β (SE)
Executive function value Polyvictimization on externalizing p 95% CI
One SD below the mean (78.09) 1.061 (.423) 0.012 [0.236–2.001]
At the mean (M = 93.11) 0.468 (.265) 0.078 [−0.074–1.011]
One SD above the mean (108.13) −0.125 (.367) 0.732 [−0.920–0.558]
Open in a new tab

Discussion

Polyvictimization was highly prevalent in the present sample, as most of the children experienced three or more discrete types of maltreatment exposure, consistent with other high-risk samples (Lau et al., 2005; Manly et al., 2001). Externalizing problems were of clinical concern for the majority of this sample (57.3%), in line with prior research that maltreated children often exhibit elevated externalizing problems (Fleckman et al., 2016). Contrary to our first hypothesis, the final model indicated that polyvictimization and externalizing problems were marginally, but not significantly, related. This finding was unexpected, given the rich literature supporting a link between polyvictimization and externalizing behaviors (e.g., Keil & Price, 2006). Yet, the interaction of polyvictimization×EF was significant. This suggests that polyvictimization, as quantified from CWS records, may be a useful way of parameterizing individual differences in maltreatment experiences. Findings indicate that the relationship between polyvictimization and externalizing problems may be contingent on underlying moderators. It is possible, given the high prevalence of externalizing behaviors in this high-risk sample, that there was insufficient variability to detect a direct link between polyvictimization and externalizing behaviors, lending support to the notion that underlying moderators are particularly valuable to delineating the association between early adverse experiences and externalizing symptomatology.

Results extend prior work which has linked maltreatment experiences with risk for the development of externalizing behavior following single victimization exposure, such as physical abuse (Lansford et al., 2002) or childhood sexual abuse (Jones et al., 2013). While informative, given the high rates of polyvicimtization within maltreated populations, it is difficult and potentially of limited clinical utility to characterize the impact of a specific maltreatment type in isolation. The current study sought to address this gap by examining the role of polyvictimization in predicting externalizing symptomatology. We endeavored to build upon the knowledge base in this area by considering the role of EF skills – deficits documented in children with externalizing behaviors (Schoemaker et al., 2013).

EF is a key mechanism of interest in maltreated youth populations as prior research has demonstrated that children in high-risk samples often exhibit lower levels of EF (Roos, Kim, Schnabler, & Fisher, 2016). Consistently, foster children in this sample had significantly lower EF skills compared to the community controls. As noted, the literature has provided support for both mediation and moderation models of EF in the context of adversity in a variety of developmental stages (Obradović, 2010; Sulik et al., 2015).

The present study did not support a mediation model, suggesting that EF does not fully account for the link between polyvictimization and externalizing symptomatology in our sample. Rather, our results supported a moderation model, indicating that individual differences in polyvictimized children’s EF skills serve as a factor in predicting externalizing problems and that only children with low EF scores were at high risk for externalizing problems following polyvictimization. Polyvictimized children with average, or above average, EF scores were buffered against externalizing problems. Findings are in line with previous research demonstrating that higher EF aptitude predicts resilient functioning following adversity; in a study of homeless youth, children’s EF emerged as the strongest predictor of adaptive functioning, including lower levels of externalizing behaviors (Obradović, 2010). One possibility is that children with higher EF had more supportive parenting, even within the context of polyvictimization, contributing to enhanced EF and better behavior regulation (i.e., lower externalizing problems). Positive parenting practices are associated with both greater EF and regulatory capacities (Bernier, Carlson, & Whipple, 2010).

Inconsistent findings in the literature in characterizing EF as a mediator or moderator between maltreatment profiles and externalizing symptoms may be due in part to several factors, such as the limited number of studies, measures of EF utilized, the operationalization of maltreatment/polyvictimization, and the developmental age group investigated. Indubitably, EF plays a significant underlying role between maltreatment, polyvictimization, and externalizing symptoms and should continue to be explored in a myriad of samples.

Implications

Externalizing problems observed at young ages place children at higher risk for future maladaptive outcomes and trajectories, including risk-taking, substance abuse, and incarceration (Campbell et al., 2000), and polyvictimization is a key risk factor. Externalizing problems are among the most reported behavioral issues in early childhood (Keenan & Wakschlag, 2004), comprising up to one half of mental health service referrals (Belsky, Melhuish, Barnes, Leyland, & Romaniuk, 2006). Without intervention, externalizing problems often remain stable and may increase from preschool across the lifespan (Stormont, 2002). Therefore, identifying young children at the highest risk of developing externalizing problems may ameliorate the likelihood of maladaptive trajectories. Our study suggests that EF skills may be useful in identifying maltreated children at the highest risk for externalizing problems.

EF may help identify children at highest risk and inform intervention design and optimization, potentially decreasing the development or exacerbation of early externalizing problems. While the research on this topic is limited, preliminary studies are promising. One intervention designed to enhance school readiness in foster care children, Kids in Transition to School (KITS), demonstrated significant positive impacts on children’s self-regulatory skills, key components related to EF (Pears et al., 2013). A separate program for socioeconomically disadvantaged children found that baseline EF skills in 4-year-old children moderated the impact of the intervention on several positive outcomes (Bierman, Nix, Greenberg, Blair, & Domitrovich, 2008). EF is also a malleable skill, potentially responsive to therapeutic techniques and interventions (Diamond & Lee, 2011).

Results support that EF can serve both as buffering and a risk factor following adversity. For example, children with low EF could be more likely than those with higher EF skills to model negative behaviors (e.g., social learning theory) or be more affected by the lack of sensitive and responsive caregiving (e.g., attachment theory), placing them at elevated risk for developing externalizing problems following polyvictimization.

Limitations

Given the nature of secondary data analysis, our study was limited to one measure of EF from a subdomain of an overall neuropsychological test and the subtests (e.g., Visual Attention) could not be computed. Future research would benefit from replicating findings using more standardized EF tasks, such as the Flanker task and dimensional change card sort (Gershon et al., 2013). However, the NEPSY and its subtests have been used as a measure of EF for preschool-aged children in other studies (e.g., Klenberg, Korkman, & Lahti-Nuuttila, 2001; Thorell & Wåh€lstedt, 2006). In the MCS, reasons necessitating out-of-home care (e.g., domestic violence) are not part of the coding system and may have been coded under other types of maltreatment (e.g., emotional neglect), potentially inflating the prevalence. Further, missing data on the NEPSY reduced the sample size for the moderation model; results were not missing at random in that children whose mothers reported lower education were less likely to have participated in the EF test. Therefore, results may be less applicable for children with lower educated caregivers. Lastly, results may be less generalizable to more racially or ethnically diverse samples. Despite these limitations, the study was adequately powered to explore the moderating role of EF in the context of polyvictimization and externalizing problems.

Future directions

Future research can build upon such findings to explore the trajectories of children at highest risk due to lower EF skills, such as conducting longitudinal analyses to explore how baseline lower EF and attention skills predict children’s well-being over time. Intervention programs that target EF skills may be critical for polyvictimized children and may mitigate the likelihood of developing externalizing problems, thus increasing the likelihood of healthier trajectories for these children across the lifespan.

Acknowledgments

Research support was provided by the following grants: R01 MH059780, NIMH, U.S. PHS; R21 MH065046; and R01 DA021424, NIDA, U.S. PHS.

Philip A. Fisher received support from NIH grants R01 HD075716 and P50 DA035763.

Leslie E. Roos received support from HHS-2014-ACF-ACYF-CA-0803.

References

  1. Achenbach TM (1991). Child Behavior Checklist/4–18 Burlington, VT: University of Vermont, Dept. of psychiatry. [Google Scholar]
  2. Afifi TO, & MacMillan HL (2011). Resilience following child maltreatment: A review of protective factors. The Canadian Journal of Psychiatry, 56(5), 266–272. doi: 10.1177/070674371105600505 [DOI] [PubMed] [Google Scholar]
  3. Barkley RA (1996). Linkages between attention and executive functions. In Lyon GR, & Krasnegor NA (Eds.), Attention, memory, and executive function (pp. pp. 307–326). Baltimore, MD: Paul H. Brookes. [Google Scholar]
  4. Barnett D, Manly JT, & Cicchetti D (1993). Defining child maltreatment: The interface between policy and research. Child Abuse, Child Development, and Social Policy, 8, 7–73. [Google Scholar]
  5. Belsky J, Melhuish E, Barnes J, Leyland AH, & Romaniuk H (2006). Effects of Sure Start local programmes on children and families: Early findings from a quasi-experimental, cross sectional study. British Medical Journal, 332(7556), 1476–1481. doi: 10.1136/bmj.38853.451748.2F [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Bernier A, Carlson SM, & Whipple N (2010). From external regulation to self-regulation: Early parenting precursors of young children’s executive functioning. Child Development, 81(1), 326–339. doi: 10.1111/cdev.2010.81.issue-1 [DOI] [PubMed] [Google Scholar]
  7. Bierman KL, Nix RL, Greenberg MT, Blair C, & Domitrovich CE (2008). Executive functions and school readiness intervention: Impact, moderation, and mediation in the Head Start REDI program. Development and Psychopathology, 20(03), 821–843. doi: 10.1017/S0954579408000394 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Campbell SB, Shaw DS, & Gilliom M (2000). Early externalizing behavior problems: Toddlers and preschoolers at risk for later maladjustment. Development and Psychopathology, 12(3), 467–488. doi: 10.1017/S0954579400003114 [DOI] [PubMed] [Google Scholar]
  9. Cecil CA, Viding E, Fearon P, Glaser D, & McCrory EJ (2017). Disentangling the mental health impact of childhood abuse and neglect. Child Abuse & Neglect, 63, 106–119. doi: 10.1016/j.chiabu.2016.11.024 [DOI] [PubMed] [Google Scholar]
  10. Cyr K, Chamberland C, Lessard G, Clément MÈ, Wemmers JA, Collin-Vézina D, … Damant D (2012). Polyvictimization in a child welfare sample of children and youths. Psychology of Violence, 2(4), 385–400. doi: 10.1037/a0028040 [DOI] [Google Scholar]
  11. Diamond A, & Lee K (2011). Interventions shown to aid executive function development in children 4 to 12 years old. Science, 333(6045), 959–964. doi: 10.1126/science.1204529 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Fay-Stammbach T, Hawes DJ, & Meredith P (2014). Parenting influences on executive function in early childhood: A review. Child Development Perspectives, 8(4), 258–264. doi: 10.1111/cdep.2014.8.issue-4 [DOI] [Google Scholar]
  13. Finkelhor D, Turner H, Hamby SL, & Ormrod R (2011). Polyvictimization: Children’s exposure to multiple types of violence, crime, and abuse. National Survey of Children’s Exposure to Violence Office of Juvenile Justice and Delinquency Prevention, US Department of Justice, Washington, DC. [Google Scholar]
  14. Fisher PA, Gunnar MR, Chamberlain P, & Reid JB (2000). Preventive intervention for maltreated preschool children: Impact on children’s behavior, neuroendocrine activity, and foster parent functioning. Journal of the American Academy of Child & Adolescent Psychiatry, 39(11), 1356–1364. doi: 10.1097/00004583-200011000-00009 [DOI] [PubMed] [Google Scholar]
  15. Fleckman JM, Drury SS, Taylor CA, & Theall KP (2016). Role of direct and indirect violence exposure on externalizing behavior in children. Journal of Urban Health, 93(3), 479–492. doi: 10.1007/s11524-016-0052-y [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Garon N, Bryson SE, & Smith IM (2008). Executive function in preschoolers: A review using an integrative framework. Psychological Bulletin, 134(1), 31–60. doi: 10.1037/0033-2909.134.1.31 [DOI] [PubMed] [Google Scholar]
  17. Gershon RC, Wagster MV, Hendrie HC, Fox NA, Cook KF, & Nowinski CJ (2013). NIH toolbox for assessment of neurological and behavioral function. Neurology, 80(11 Supplement 3), S2–S6. doi: 10.1212/WNL.0b013e3182872e5f [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Greeson JK, Briggs EC, Kisiel CL, Layne CM, Ake GS III, Ko SJ, … Fairbank JA (2011). Complex trauma and mental health in children and adolescents placed in foster care: Findings from the National Child Traumatic Stress Network. Child Welfare, 90(6), 91–108. [PubMed] [Google Scholar]
  19. Hayes AF (2013). Introduction to mediation, moderation, and conditional process analysis: A regression-based approach New York, NY: Guilford Press. [Google Scholar]
  20. Jones DJ, Lewis T, Litrownik A, Thompson R, Proctor LJ, Isbell P, … Runyan D (2013). Linking childhood sexual abuse and early adolescent risk behavior: The intervening role of internalizing and externalizing problems. Journal of Abnormal Child Psychology, 41(1), 139–150. doi: 10.1007/s10802-012-9656-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Kane M, & Engle R (2003). Working-memory capacity and the control of attention: The contributions of goal neglect, response competition, and task set to Stroop interference. Journal of Experimental Psychology: General, 132(1), 47–70. doi: 10.1037/0096-3445.132.1.47 [DOI] [PubMed] [Google Scholar]
  22. Keenan K, & Wakschlag LS (2004). Are oppositional defiant and conduct disorder symptoms normative behaviors in preschoolers? A comparison of referred and nonreferred children. American Journal of Psychiatry, 161(2), 356–358. doi: 10.1176/appi.ajp.161.2.356 [DOI] [PubMed] [Google Scholar]
  23. Keil V, & Price JM (2006). Externalizing behavior disorders in child welfare settings: Definition, prevalence, and implications for assessment and treatment. Children and Youth Services Review, 28(7), 761–779. doi: 10.1016/j.childyouth.2005.08.006 [DOI] [Google Scholar]
  24. Kim J, & Cicchetti D (2010). Longitudinal pathways linking child maltreatment, emotion regulation, peer relations, and psychopathology. Journal of Child Psychology and Psychiatry, 51(6), 706–716. doi: 10.1111/j.1469-7610.2009.02202.x [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Klenberg L, Korkman M, & Lahti-Nuuttila P (2001). Differential development of attention and executive functions in 3-to 12-year-old Finnish children. Developmental Neuropsychology, 20(1), 407–428. doi: 10.1207/S15326942DN2001_6 [DOI] [PubMed] [Google Scholar]
  26. Kochanska G, Murray KT, & Harlan ET (2000). Effortful control in early childhood: Continuity and change, antecedents, and implications for social development. Developmental Psychology, 36(2), 220–232. doi: 10.1037/0012-1649.36.2.220 [DOI] [PubMed] [Google Scholar]
  27. Korkman M, Kemp S, & Kirk U (1998). NEPSY: A developmental neuropsychological assessment San Antonio, TX: Psychological Corporation. [Google Scholar]
  28. Lansford JE, Dodge KA, Pettit GS, Bates JE, Crozier J, & Kaplow J (2002). A 12-year prospective study of the long-term effects of early child physical maltreatment on psychological, behavioral, and academic problems in adolescence. Archives of Pediatrics & Adolescent Medicine, 156(8), 824–830. doi: 10.1001/archpedi.156.8.824 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Lansford JE, Malone PS, Stevens KI, Dodge KA, Bates JE, & Pettit GS (2006). Developmental trajectories of externalizing and internalizing behaviors: Factors underlying resilience in physically abused children. Development and Psychopathology, 18(1), 35–55. doi: 10.1017/S0954579406060032 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Lau AS, Leeb RT, English D, Graham JC, Briggs EC, Brody KE, & Marshall JM (2005). What’s in a name? A comparison of methods for classifying predominant type of maltreatment. Child Abuse & Neglect, 29(5), 533–551. doi: 10.1016/j.chiabu.2003.05.005 [DOI] [PubMed] [Google Scholar]
  31. Leeb RT, Lewis T, & Zolotor AJ (2011). A review of physical and mental health consequences of child abuse and neglect and implications for practice. American Journal of Lifestyle Medicine, 55, 454–468. doi: 10.1177/1559827611410266 [DOI] [Google Scholar]
  32. Manly JT, Kim JE, Rogosch FA, & Cicchetti D (2001). Dimensions of child maltreatment and children’s adjustment: Contributions of developmental timing and subtype. Development and Psychopathology, 13(04), 759–782. [PubMed] [Google Scholar]
  33. Masten AS (2001). Ordinary magic: Resilience processes in development. American Psychologist, 56(3), 227. doi: 10.1037/0003-066X.56.3.227 [DOI] [PubMed] [Google Scholar]
  34. Miller SE, & Marcovitch S (2015). Examining executive function in the second year of life: Coherence, stability, and relations to joint attention and language. Developmental Psychology, 51(1), 101–114. doi: 10.1037/a0038359 [DOI] [PubMed] [Google Scholar]
  35. Moylan CA, Herrenkohl TI, Sousa C, Tajima EA, Herrenkohl RC, & Russo MJ (2010). The effects of child abuse and exposure to domestic violence on adolescent internalizing and externalizing behavior problems. Journal of Family Violence, 25(1), 53– 63. doi: 10.1007/s10896-009-9269-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Obradović J (2010). Effortful control and adaptive functioning of homeless children: Variable-focused and person-focused analyses. Journal of Applied Developmental Psychology, 31(2), 109–117. doi: 10.1016/j.appdev.2009.09.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Olson SL, Tardif TZ, Miller A, Felt B, Grabell AS, Kessler D, … Hirabayashi H (2011). Inhibitory control and harsh discipline as predictors of externalizing problems in young children: A comparative study of US, Chinese, and Japanese preschoolers. Journal of Abnormal Child Psychology, 39(8), 1163–1175. doi: 10.1007/s10802-011-9531-5 [DOI] [PubMed] [Google Scholar]
  38. Patterson GR, & Yoerger K (2002). A developmental model for early-and late-onset delinquency. In Reid JB, Snyder J, & Patterson GR (Eds.), Antisocial behavior in children and adolescents: A developmental analysis and model for intervention (pp. pp. 147–172). Washington, DC: American Psychological Association. [Google Scholar]
  39. Pears KC, Fisher PA, Kim HK, Bruce J, Healey CV, & Yoerger K (2013). Immediate effects of a school readiness intervention for children in foster care. Early Education & Development, 24(6), 771–791. doi: 10.1080/10409289.2013.736037 [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Pears KC, Kim HK, & Fisher PA (2008). Psychosocial and cognitive functioning of children with specific profiles of maltreatment. Child Abuse & Neglect, 32(10), 958–971. doi: 10.1016/j.chiabu.2007.12.009 [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Roman GD, Ensor R, & Hughes C (2016). Does executive function mediate the path from mothers’ depressive symptoms to young children’s problem behaviors? Journal of Experimental Child Psychology, 142, 158–170. doi: 10.1016/j.jecp.2015.09.022 [DOI] [PubMed] [Google Scholar]
  42. Roos LE, Kim HK, Schnabler S, & Fisher PA (2016). Children’s executive function in a CPS-involved sample: Effects of cumulative adversity and specific types of adversity. Children and Youth Services Review, 71, 184–190. doi: 10.1016/j.childyouth.2016.11.008 [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Schoemaker K, Mulder H, Deković M, & Matthys W (2013). Executive functions in preschool children with externalizing behavior problems: A meta-analysis. Journal of Abnormal Child Psychology, 41(3), 457–471. doi: 10.1007/s10802-012-9684-x [DOI] [PubMed] [Google Scholar]
  44. Slade EP, & Wissow LS (2007). The influence of childhood maltreatment on adolescents’ academic performance. Economics of Education Review, 26(5), 604–614. doi: 10.1016/j.econedurev.2006.10.003 [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Spinrad TL, Eisenberg N, Gaertner B, Popp T, Smith CL, Kupfer A, … Hofer C (2007). Relations of maternal socialization and toddlers’ effortful control to children’s adjustment and social competence. Developmental Psychology, 43(5), 1170–1186. doi: 10.1037/0012-1649.43.5.1170 [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Stein E, Evans B, Mazumdar R, & Rae-Grant N (1996). The mental health of children in foster care: A comparison with community and clinical samples. The Canadian Journal of Psychiatry, 41(6), 385–391. doi: 10.1177/070674379604100610 [DOI] [PubMed] [Google Scholar]
  47. Stormont M (2002). Externalizing behavior problems in young children: Contributing factors and early intervention. Psychology in the Schools, 39(2), 127–138. doi: 10.1002/(ISSN)1520-6807 [DOI] [Google Scholar]
  48. Stovall-Mcclough KC, & Dozier M (2004). Forming attachments in foster care: Infant attachment behaviors during the first 2 months of placement. Development and Psychopathology, 16(02), 253–271. doi: 10.1017/S0954579404044505 [DOI] [PubMed] [Google Scholar]
  49. Sulik MJ, Blair C, Mills-Koonce R, Berry D, & Greenberg M (2015). Early parenting and the development of externalizing behavior problems: Longitudinal mediation through children’s executive function. Child Development, 86(5), 1588–1603. doi: 10.1111/cdev.2015.86.issue-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Thorell LB, & Wåhlstedt C (2006). Executive functioning deficits in relation to symptoms of ADHD and/or ODD in preschool children. Infant and Child Development, 15(5), 503– 518. doi: 10.1002/(ISSN)1522-7219 [DOI] [Google Scholar]
  51. Turner HA, Finkelhor D, & Ormrod R (2010). Poly-victimization in a national sample of children and youth. American Journal of Preventive Medicine, 38(3), 323–330. doi: 10.1016/j.amepre.2009.11.012 [DOI] [PubMed] [Google Scholar]

RESOURCES