Child maltreatment and emotion socialization: Associations with executive function in the preschool years
Introduction
Executive function (EF) can be broadly conceptualized as a set of interrelated, higher-order cognitive processes that enable individuals to direct their attention, thoughts and actions for adaptive reasoning, problem-solving and planning, as indexed by tests of working memory, inhibitory control and cognitive flexibility (Diamond, 2013). Impairments in EF during early childhood have been associated with increased risk for a host of later problems pertaining to socioemotional competence and externalizing problems (Carlson, Zelazo, & Faja, 2013). With respect to cognitive outcomes, early childhood measures of EF predict school readiness, literacy, and mathematic abilities more consistently than IQ or early achievement scores (Clark, Sheffield, Wiebe, & Espy, 2013). Longitudinal research has likewise found early childhood deficits in self-control − a related aspect of EF − to predict social adjustment in adulthood (Moffitt et al., 2011). Such evidence points to EF as a high-priority target for early intervention and prevention, and has emphasized the need to identify risk and protective factors associated with its development (Ursache, Blair, & Raver, 2012).
Due to the prolonged development of the prefrontal cortex system, its structure and function are shaped by environmental conditions and social experiences that can be both positive and enriching (e.g., stimulation, responsive caregiving) or negative and disruptive (e.g., maltreatment, poverty), particularly during periods of rapid neural growth or plasticity such as early childhood (Kolb et al., 2012). According to the neurobiological models of early stress, exposure to aggression, threat or neglect can produce dysregulated stress responses (e.g., allostatic load) leading to increased states of emotional over-arousal (e.g., hypervigilance, hyperactivity, numbing) and disruptions to volitional attention and EF (De Bellis, 2005; McCrory, De Brito, & Viding, 2010).
Research into the influences of family environment on emerging EF has focused largely on the periods of middle-to-late childhood, and on the extreme disturbances in caregiving characteristic of child maltreatment. Exposure to maltreatment in these periods has been associated with deficits in processing speed, inhibition, auditory and working memory, abstract thinking, and attention (Beers & De Bellis, 2002; DePrince, Weinzierl, & Combs, 2009; Mezzacappa, Kindlon, & Earls, 2001). It has likewise been associated with poor spatial working memory and everyday memory (Augusti & Melinder, 2013; Moradi, Doost, Taghavi, Yule, & Dalgleish, 1999).
Considerably less is known about the influences of maltreatment on EF during early childhood. Two studies of young children in foster care have found them to perform worse on tests of inhibitory control than non-foster children (Lewis, Dozier, Ackerman, & Sepulveda-Kozakowski, 2007) Pears et al., 2010), whereas one found them comparable (Pears & Fisher, 2005). Findings regarding children under the care of their biological parents have also been mixed. Physical abuse has been associated with poor parent-reported EF among children (aged 4–7 years) known to child protection services (Kim-Spoon, Haskett, Longo, & Nice, 2012). However, other research has found no differences between maltreated and non-maltreated preschoolers on measures of inhibitory control Cipriano-Essel, Skowron, Stifter, & Teti (2013). Such findings highlight the need for ongoing research into the influences of maltreatment on EF in early childhood, and raise questions regarding other family-level processes that may amplify or protect against such influences during this period. Importantly, there is growing evidence to indicate that not only extreme environmental insults such as maltreatment, but quality of caregiving in common parenting domains, accounts for individual differences in emerging EF.
As identified in a recent systematic review, the parenting variables that have been researched most extensively in this regard are those related to sensitivity (Fay-Stammbach, Hawes, & Meredith, 2014). As conceptualized in attachment theory, sensitive caregiving (e.g., positive affect, warmth, absence of hostility) is assumed to promote the internalization of regulatory strategies. Findings from longitudinal research across infancy and early childhood have been largely consistent is demonstrating associations between high levels of sensitivity (and low levels of hostility) and emerging EF in early childhood (e.g., Blair et al., 2001, Cuevas et al., 2014, Kraybill and Bell, 2013; Rhoades, Greenberg, Lanza, & Blair, 2011). Likewise, a randomized controlled trial of an attachment-based parenting program found it to enhance EF (cognitive flexibility) in maltreated foster children (Lewis-Morrarty, Dozier, Bernard, Terracciano, & Moore, 2012).
Traditionally, developmental research into sensitive caregiving has focused largely on the period of infancy, during which time it has often been operationalized broadly in terms of frequency and appropriateness of mothers’ responses to infants’ emotional displays (e.g., Cohn and Tronick, 1983, Field, 1981). In recent years researchers have focused increasingly on developmental periods beyond infancy, and have conceptualized sensitive caregiving in terms of specific emotion socialization practices comprising supportive versus nonsupportive parental reactions to children’s emotional reactions. According to Eisenberg, Cumberland, and Spinrad (1998), these emotion-related socialization behaviors (ERSBs) largely fall into four domains: 1) parents’ reactions to children’s emotions; 2) parents’ discussion of emotion; 3) parents’ expression of emotion; and 4) parents’ selection or modification of the child’s situation. The potential to precisely operationalize and target these ERSBs in skill-based parenting interventions can be seen as an important advantage over more global conceptualizations of caregiver sensitivity.
Research to date has associated ERSBs with behavioral and self-regulatory competencies in a range of populations, including typically developing children (e.g., Cole, Dennis, Smith-Simon, & Cohen, 2009) and those with clinic-referred difficulties (e.g., Pasalich, Waschbusch, Dadds, Hawes, 2014).. Such research with families characterized by maltreatment has been particularly limited, yet has nonetheless indicated that ERSBs may be important to regulatory outcomes among such children. Most notably, such research has found that among physically maltreating mothers, use of ERSBs mediate the relationship between maltreatment and regulation of emotional expression and parent/child-reports of emotional arousal and regulation in children aged 6-to-12 years (Shipman and Zeman, 2001, Shipman et al., 2007). This evidence of interplay between maltreatment status and ERSBs raises the question of whether ERSBs may likewise protect against the adverse effects that maltreatment appears to have on emerging EF (i.e., cognitive control) in early childhood. Interestingly, these studies indicate that associations between maltreatment status and ERSBs (e.g., emotion coaching) may be only moderate in size (Shipman et al., 2007). As such, while parents at risk for maltreatment are by definition characterized by negative parenting practices, considerable variability in levels of positive and negative ERSBs may nonetheless be seen among such parents.
The major aim of this study was to examine the relationship between maltreatment and EF in preschool-aged children. Although a range of parameters can be used to index magnitude of maltreatment, there is growing evidence to indicate that most victims of childhood maltreatment experience multiple forms of victimization, and that this multiplicity may be particularly influential in conferring risk for negative biopsychosocial outcomes (Chartier, Walker, & Naimark, 2010; Edwards et al., 2014); Finkelhor, Ormrod, Turner, & Hamby, 2005). In line with this, we examined maltreatment in terms of the number of maltreatment types to which children had been exposed.
Based on this approach we examined the unique contributions of maltreatment and ERSBs to EF, and the interplay between maltreatment and ERSBs in explaining EF. As outlined, considerably less is known about the influences of maltreatment on cognitive development in early childhood compared to later periods, and EF can be considered a particularly critical domain of such development (Fay-Stammbach et al., 2014). Moreover, although previous research has indicated that ERSBs may in part determine the emotion regulation outcomes of children exposed to maltreatment, we are aware of no studies in which ERSBs have been examined in relation to EF. Given evidence to suggest that early executive attention may contribute to later emotion regulation (Rothbart, Sheese, Rueda, & Posner, 2011), a specific focus on EF is well-justified. Importantly, the current study was designed to investigate EF using performance-based indices, thereby building on previous research into children’s emotion regulation that has relied on self-report measurement (Shipman and Zeman, 2001, Shipman et al., 2007).
It was hypothesized, first, that exposure to maltreatment would be associated with poor EF. A noteworthy limitation of much previous research has been a focus on maltreatment in isolation from broader ecological risk factors. There is emerging evidence that both distal, demographic (e.g., low socioeconomic status, ethnicity) and more proximal risk factors (e.g., parental depression) are negatively associated with EF development (e.g., Hughes, Roman, Hart, & Ensor, 2013; Rhoades et al., 2011). Furthermore risk factors related to parental depression, low-income, and ethnicity have been found to co-occur in families of maltreated children (Cicchetti, 2012; Haskett, Allaire, Kreig, & Hart, 2008). In order to attribute findings to variables of theoretical interest, we examined whether maltreatment would be associated with EF independent of ecological risk factors (e.g., ethnicity, maternal education and depression). Likewise, this association was predicted to be independent of developmental child factors that have been found to covary with EF, including child receptive language and bilingualism.
Second, it was hypothesized that quality of ERSBs would moderate the association between maltreatment and EF. As noted, previous research has found quality of parenting to mediate the effects of maltreatment on emotion regulation in older children (Shipman and Zeman, 2001, Shipman et al., 2007). Such a relationship assumes that maltreatment is directly associated with individual differences in parental sensitivity, however, numerous other studies have found this not to be the case. For example, after controlling for covariates including maternal education, and child age and IQ, Cipriano-Essel et al. (2013) found no differences in observations of maternal warmth/hostility between maltreatment and non-maltreatment children. Given the lack of clarity regarding direct associations between maltreatment and parental sensitivity, interplay between maltreatment and ERSBs was examined in terms of moderation rather than mediation. Specifically, it was predicted that high levels of supportive ERPBs (expressive/encouraging, emotion or problem-focused responses) would mitigate the relationship between maltreatment and EF, whereas high levels of nonsupportive parental responses (punitive or dismissing) would amplify this relationship.
Section snippets
Participants
Participants were children (n = 107; 39% female) aged 4 and 5 years (M = 4.75, SD = 0.57) and their caregivers. In order to examine correlates of maltreatment across various levels of severity, recruitment targeted both families involved with child protection services as well as those with no such involvement. The sample comprised n = 58 at-risk/maltreated children, recruited through an early intervention program to which families were referred by child protection services based on high risk for
Descriptive analyses
Means and correlations among the key child and family variables are reported in Table 1. Child age, ethnicity and bilingualism were all significantly correlated with the EF composite index. Overall EF was also significantly correlated with maternal education, and depression. Neither receptive language, nor the socio-demographic variables of parental age, child sex, single parent status, or parental source of income (receipt of welfare payments), were significantly related to EF at the bivariate
Discussion
This study examined associations between family environment and EF in the preschool years, a period during which self-regulatory capacities undergo significant and rapid development. Of particular interest was the relationship between exposure to maltreatment and EF, and the potential for caregiver emotion socialization to moderate this relationship. Consistent with the study predictions, exposure to substantiated maltreatment was found to be associated with reduced capacity for EF.
References (72)
- et al.
Commentary on the review of measures of early childhood social and emotional development: Conceptualization, critique, and recommendations
Journal of Applied Developmental Psychology
(2016) - et al.
Inhibitory control and emotion regulation in preschool children
Cognitive Development
(2007) - et al.
Separate and cumulative effects of adverse childhood experience in predicting adult health care utilization
Child Abuse and Neglect
(2010) - et al.
Executive function performance and trauma exposure in a community sample of children
Child Abuse & Neglect
(2009) - et al.
Measuring poly- victimization using the juvenile victimization questionnaire
Child Abuse & Neglect
(2005) - et al.
Protective and vulnerability factors for physically abused children: effects of ethnicity and parenting context
Child Abuse & Neglec
(2008) - et al.
Longitudinal study of self-regulation, positive parenting, and adjustment problems among physically abused children
Child Abuse & Neglect
(2012) - et al.
Cognitive flexibility and theory of mind outcomes among foster children: Preschool follow-up results of a randomized clinical trial
Journal of Adolescent Health
(2012) - et al.
Examining the association between parenting and childhood depression: A meta-analysis
Clinical Psychology Review
(2007) - et al.
The contribution of inhibitory control to preschoolers' social-emotional competence
Journal of Applied Developmental Psychology
(2009)