From the Cradle to the Grave: The Effect of Adverse Caregiving Environments on Attachment and Relationships Throughout the Lifespan

Abstract

This paper reviews research examining the effects of adverse early caregiving on relationships throughout the lifespan. Central attachment constructs are summarized and integrated into a review of research on the long-term effects of institutional rearing and child maltreatment. Findings are interpreted within the organizational perspective on development, which conceptualizes attachment as a stage-salient task of infancy that influences the reorganization of adaptive/maladaptive functioning around subsequent stage-salient tasks. Children who experience adverse early caregiving are more likely to exhibit aberrant attachment behaviors, deficits in social-emotional competencies, and persisting difficulties in social functioning and relationship outcomes. Disorganized attachment behavior stemming from adverse early caregiving has been a major focus of this work. Intervention efforts that target mental representations related to attachment relationships can facilitate improved social functioning. Clinical implications of this work are discussed.

Relationships provide a fundamental context for development across the lifespan. These relationships include those with parents, early care and education providers, peers, and romantic partners. Because early caregiving experiences critically shape brain and behavior development, children who experience absent or abusive caregiving are denied opportunities for healthy development. Instead, adverse caregiving environments can set into motion a probabilistic developmental pathway that is characterized by increased risk for maladaptive behavior and relationship disadvantage across the lifespan.

This paper reviews a growing literature examining the association among childhood adversity, attachment disruptions, and relationships in childhood, adolescence, and adulthood. We focus on two types of childhood adversity: institutional rearing and child maltreatment. First, we review attachment theory, the central conceptual framework guiding work in this area. Next, we summarize recent empirical research and intervention efforts applying attachment constructs to the study of developmental pathways and phenomena linking extreme, pathogenic caregiving environments to social functioning and relationship outcomes later in life. Finally, we discuss the clinical implications of these findings, as well as promising directions for future work. Our approach is informed by an organizational perspective on development (Cicchetti & Toth, 2015; Sroufe, Egeland, Carlson, & Collins, 2005), which conceptualizes development as a series of functional reorganizations around stage-salient tasks. Importantly, from this perspective, the meaning of social behaviors, and not merely their occurrence, is informative. Additionally, an organizational perspective considers how stage-salient tasks may provide an opportunity for growth, resilience, and recovery, in addition to vulnerability. For these reasons, this perspective has important clinical value for researchers examining processes and pathways linking adverse early caregiving to long-term social functioning. Assessing what has been learned from this research is important, as an understanding of how childhood adversity influences attachment and relationships across the lifespan has broad implications for clinicians, researchers, policy makers, and the public at large.

Central Tenets of Attachment Theory

According to attachment theory, early caregiving experiences influence adaptation and maladaptation across the lifespan by organizing individual and relational developmental processes “from the cradle to the grave” (Bowlby, 1969/1982; p. 208). This conceptual framework has advanced our understanding of normal and abnormal development by generating testable hypotheses about the developmental processes and pathways linking early caregiving experiences to long-term outcomes. The central assumption of attachment theory is that relationships function to increase survival. Attachment theory contends that infants are biologically predisposed to bond with caregivers, signal for their protection and care, and seek proximity to those who can provide safety and support. From early interaction patterns with caregivers, infants are thought to develop an internal working model of attachment, or a set of mental representations of the self, the caregiver, and the world (Sherman, Rice, & Cassidy, 2015). These internal working models operate across the lifespan to promote survival by guiding the individual's expectations and explorations of the social and physical environment.

In addition to promoting survival, attachment relationships can propel infants along developmental pathways of adaptation or maladaptation. Ainsworth and colleagues (1978) generated significant interest in this area by bringing attention to the individual differences in the infant-caregiver attachment relationship (Ainsworth, Blehar, Waters, & Wall, 1978). Considerable research shows that a caregiver's behavior toward the child is the key determinant of the attachment relationship and the child's internal working model (de Wolff & van IJzendoorn, 1997), and that both the caregiver's behavior and the child's internal working model subsequently influence the child's social and emotional development (e.g., Sroufe, 2005). Specifically, caregivers can promote a secure attachment relationship when they are available, sensitive to a child's needs, and responsive to a child's cues. Children who are securely attached learn they can rely on the caregiver for protection, and consequently these children perceive themselves as worthy of receiving care and love, and relationships as safe and worth pursuing. In contrast, caregivers who are unavailable, low in sensitivity, or inconsistently or inappropriately responsive to a child's needs promote an insecure attachment relationship. These children may develop ambivalent expectations about their own self-worth and view relationships as hurtful, rejecting, or unsafe. Thus, early caregiving experiences can initiate cascading and reciprocal effects among attachment relationships, social development, and social adaptation over the lifespan. Importantly, research shows that secure vs. insecure attachment relationships cannot be explained by genetic factors, implying a process of environmental transmission via parent-child social interactions.

The profound and persisting effects of early caregiving on social functioning can be understood from the organizational perspective on development (see Figure 1). This perspective conceptualizes development as a series of functional reorganizations around stage-salient tasks. Importantly, the organizational perspective on development holds that the meaning of a behavior, not merely its occurrence, must be considered. For example, disengaging from a social conflict may be adaptive or maladaptive, depending on the social situation. Accordingly, this perspective can offer clinical insights on relationship outcomes linked to adverse caregiving, rather than only reporting incidence rates. Additionally, by focusing on the meaning of behavior, this perspective brings attention to the functions and context sensitivity of underlying behavioral and biological systems that organize attachment and relational behavior at each stage of development. Thus, this perspective considers mechanisms and mediators of early child adversity at multiple levels of analyses: biological, cognitive, behavioral, and social. Finally, the organizational perspective on development considers how each stage-salient task involves opportunities for growth, recovery, resilience, the consolidation of functional skills, as well as challenges that may open windows of vulnerability. However, overall adaptation across the lifespan is depicted as cumulative and hierarchical, such that successful resolution of an early stage-salient task increases the likelihood of subsequent successful task resolution. Thus, an organizational perspective on development is needed to comprehend the continuity and discontinuity of attachment and relational behaviors within the context of developmental changes, as well as the opportunity for resilience following early adverse experiences.

Figure 1.

Conceptual model depicting the organizational perspective on social development in the context of adverse early caregiving. This perspective suggests that failure of the caregiving environment to support healthy development around stage-salient attachment tasks may lead to deficits in social-emotional competence in early childhood that increase the likelihood for difficulties and deficits in subsequent stage-salient tasks across the lifespan. Although longitudinal studies provide support for this cumulative, hierarchical depiction of social maladaptation following early adversity, intervention studies targeting child attachment behaviors or mental representations directly or indirectly also suggest that social-emotional development can be reorganized at different stages in development, allowing for opportunities of recovery.

The literature provides strong support for the cumulative and hierarchical depiction of social adaption/maladaption across development following differences in early caregiving experiences (see Groh et al., 2014). Specifically, research shows that stage-salient attachment behaviors of infancy, such as smiling, clinging, and checking back with a caregiver, are associated with the successful resolution of subsequent relational stage-salient tasks, such as exploring peer interactions, interpreting social cues of peers, and establishing secure peer relationships. Longitudinal studies, more frequently existing with study samples experiencing typically occurring variations in early caregiving experiences, show that social and emotional competency in childhood feed forward over time to facilitate successful resolution of stage-salient tasks related to adolescent and adult peer and romantic relationships, such as establishing positive self-esteem and developing conflict resolution skills (see Grossmann, Grossmann, & Waters, 2006). It should be noted that the organizational perspective on development is inherent to the approach of many attachment researchers, though it may not be explicitly acknowledged.

Early Adversity and Attachment Disturbances in Childhood

Adverse caregiving environments, such as those characterized by neglect, deprivation, or abuse, can be viewed as extreme, pathogenic forms of insensitive caregiving within the continuum of caretaking casualty (see Sameroff & Chandler, 1975). These adverse environments deviate from typical caregiving environments such that infants experience the absence of necessary input, as well as exposure to adverse input. In institutional settings, caregiving generally is non-individualized, routine, and cursory, rather than affectionate, sensitive and responsive (van IJzendoorn et al., 2011). Similarly, children who experience maltreatment are exposed to social-emotional deprivation and/or abuse. As a result, children who experience institutional care or maltreatment are denied vital opportunities to form healthy attachments, resulting in behavior presenting as attachment disturbances and disorders, and difficulties establishing peer relationships.

Disorganized attachment, comprised of fearful, conflicted, and disoriented behaviors, has been a primary focus of empirical and clinical work on attachment disturbances related to these two forms of adverse early caregiving. Numerous independent studies have shown that frightening or unpredictable caregiver behavior (e.g., verbally or physically intrusive or hostile behavior), as opposed to insensitive behavior (e.g., missing or ignoring child's cues or emotions) promotes disorganized child attachment behaviors (e.g., freezing, hitting the parent, or contradictory behavior like running up and then pulling away) (see Lyons-Ruth & Jacobvitz, 2008). Accordingly, disorganized behaviors are thought to reflect the absence or breakdown of children's relational strategies for dealing with threat, and suggest that children do not have the opportunity to learn effective self-soothing or self-regulation skills. From an organizational perspective of development, the proposed meaning of these behaviors has important implications for the sequelae of early adverse caregiving. Meta-analyses suggest that approximately 15% of infants in normative samples exhibit disorganized attachment behavior; in contrast, the prevalence rate is two to three times higher among infants who have experienced institutional rearing (Lionetti, Pastore, & Barone, 2015; van den Dries, Juffer, van IJzendoorn, & Bakermans-Kranenburg, 2009) and up to 80-90% in child maltreatment samples (see Cicchetti & Toth, 2015; Cyr, Euser, Bakermans-Kranenburg, & van IJzendoorn, 2010). These analyses suggest that children who experience institutional care or maltreatment have a relatively equivalent chance of developing an insecure or disorganized attachment, and a relatively minor chance of developing a secure attachment.

Extreme, pathogenic caregiving environments also are associated with two attachment disorders: reactive attachment disorder (RAD), an emotionally withdrawn/inhibited pattern of behavior, and disinhibited social engagement disorder (DSED), an indiscriminately social/disinhibited pattern of behavior (see Zeanah & Gleason, 2015). Importantly, disorganized attachment is conceptualized as reflecting a state of distorted attachment, while attachment disorders are conceptualized as reflecting “non-attachment.” While the majority of children who experience institutional rearing and maltreatment exhibit insecure or disorganized attachments, and may show symptoms of RAD or DSED, most do not meet criteria for these disorders (Zeanah et al., 2015). Recent research shows RAD and DSED both arise from inadequate caregiving environments characterized by social neglect, and have distinct presentations, correlates, and response to intervention.

The core features of RAD include a general absence of attachment behaviors, including failure to seek and respond to comfort from a caregiver when distressed, as well as reduced social and emotional reciprocity with a caregiver. RAD is uncommon, and is estimated to occur in approximately 10% of children who experience inadequate caregiving (American Psychiatric Association [APA], 2013). Notably, RAD symptoms are greater in children who have experienced more severe social neglect, and dissipate rapidly after intervention efforts provide children the opportunity to form selective attachments with nurturing foster or adoptive parents (Zeanah et al., 2015). In contrast, the core features of DSED include a general absence of developmentally appropriate reticence around unfamiliar adults, and a failure to check in and maintain proximity with caregivers. Importantly, children who meet criteria for DSED may exhibit disinhibited behavior with strangers and show typical attachment behaviors with primary caregivers. Approximately 20% of children who have experienced social neglect meet criteria for DSED (APA, 2013). Numerous adoption studies show DSED symptoms are associated with length of time in institutional care, persist after intervention, and are not associated with the quality of caregiving in foster or adoptive homes (Zeanah et al., 2015).

Although post-institutionalized (PI) children show remarkable recovery after they are placed in nurturing families, there is considerable variability in developmental outcomes. As a group, PI children exhibit delays and deficits in identifying and interpreting emotions, perspective taking, understanding and observing social boundaries, and regulating emotions (Gunnar & Quevedo, 2007). From an organization perspective of development, failure to resolve these stage-salient tasks put PI children at increased risk for persisting social competency deficits and relationship difficulties. Overall, research shows social problems are positively correlated with age at adoption, with variability related to type of social outcome and severity of early deprivation (Zeanah, Gunnar, McCall, Kreppner, & Fox, 2011). Interestingly, children adopted before turning one years old exhibit attachment security with adoptive parents that is comparable to non-adopted peers; in contrast, adoption after 12 months is associated with increased risk for insecure, disorganized, and atypical attachment patterns (van den Dries et al., 2009). Recent research shows that nearly all PI infants, regardless of age at adoption, form some kind of attachment with adoptive parents by 9 months post-placement (Carlson, Hostinar, Mliner, & Gunnar, 2014). These findings speak to the robust capacity for human attachment.

Consistent with an organizational perspective on development, research shows the negative relational patterns acquired in a maltreating early caregiving environment become incorporated into the child's mental representations of self, others, and the social world, which are relevant for later peer relations (Cicchetti & Toth, 2015). Accordingly, maltreatment may lead to negative expectations of the self and others, as well as the concept of relationships as involving victimization and coercion. Through these disturbances in representational models, the negative effects of early maltreatment can be carried forward to later stage-salient developmental tasks, such as establishing peer relationships. This conceptualization of continuity in relational functioning is supported by a large body of research showing maltreatment is associated with later peer difficulties in childhood and adolescence (see Riggs, 2010).

In comparison to their peers, children who have experienced maltreatment are more likely to actively withdraw from social interactions and display more aggressive behavior when they do interact (Cicchetti & Toth, 2015). This conflicted pattern of both “flight” and “fight” interaction with peers is consistent with research showing that maltreatment is associated with disorganized attachment behavior. It suggests children who have experienced maltreatment may operate from a disorganized representational model that includes elements of both victim and victimizer (Cicchetti & Toth, 2015). Accordingly, research shows that maltreated children are more likely than nonmaltreated children to bully other children and more likely to be victimized by their peers (Cicchetti & Toth, 2015). Additionally, child maltreatment, particularly child sexual abuse, has been associated with age-inappropriate sexualized behavior in children (Trickett, Noll, & Putnam, 2011). Maltreatment also places children at risk for being rejected and victimized by their peers. Research shows that the continuation of maltreatment is associated with repeated (as opposed to transitory) peer rejection, and that aggression largely accounts for the association between maltreatment and peer rejection (Bolger & Patterson, 2001). Moreover, genetically informed twin designs show that early experiences of maltreatment play a causal role in the development of antisocial behavior like aggression, independent of genetic effects (Jaffee, Caspi, Moffitt, & Taylor, 2004). Importantly, these results suggest that preventing maltreatment experiences can prevent its later impact on peer relationships.

Additionally, research shows that the effects of maltreatment on disrupted peer functioning may be accounted for by perturbations in cognitive and emotional processes (Cicchetti & Toth, 2015). In response to a research paradigm that involves two adult experimenters acting out a social conflict scenario that elicits simulated anger in both the actors, children who have experienced maltreatment demonstrate patterns of both overcontrolled (i.e., low reactivity, high avoidance) and undercontrolled (i.e., both positive and negative affect, disorganized behavior, and mismatched social cognitions) physiological and behavioral regulatory profiles (Maughan & Cicchetti, 2002). This conflicted pattern of regulation is again akin to a disorganized attachment profile. It suggests children who have experienced maltreatment may fail to develop health regulatory strategies, and instead may learn different strategies of coping with social adversity that are adaptive in early environments, and maladaptive in later peer settings. Children who have experienced physical abuse display deficits in social information processing, such as making errors in encoding social cues, exhibiting a bias toward attributing hostile intent, generating more aggressive responses, and positively evaluating aggression as an appropriate response (Cicchetti et al., 2015). These deficits mediate the association between physical abuse and aggression towards peers.

Early Adversity and Attachment Disturbances in Adolescence

Although several longitudinal studies have shown that PI youth have higher social problems in adolescence compared to non-adopted youth, some studies have found no group differences (Hawk & McCall, 2010). Numerous studies have examined parent and teacher report of peer difficulties, and shown that PI adolescents have higher rates compared to non-adopted youth peer problems (e.g., Gunnar & Quevedo, 2007). Longitudinal data from the English and Romanian Adoptees Study show that adoption after six months is associated with symptoms of DSED persisting to age 15; in contrast, children whose exposure to institutional care was limited to six months did not show persisting effects (Rutter et al., 2010). These findings suggest a sensitive period of early neurobehavioral development that is significantly influenced by the quality of caregiving may open at six months old. Since DSED behaviors transgress social boundaries and cues, it is not surprising that youth with these problems report difficulty establishing meaningful peer relationships (Kay & Green, 2013).

Longitudinal research shows that disruption to attachment behaviors in infancy, stemming from childhood adversity, confer risk for social maladaptation in adolescence. In social contexts, such as peer interaction, aberrant or absent attachment behaviors are likely to negatively affect the “readability” of the youth's social signals, as well as the reactions the teen can elicit from other relationship partners. During adolescence, age-appropriate behavior related to emerging sexual identify is a stage-salient task. Longitudinal research shows that child maltreatment, particularly sexual abuse, is associated with maladaptive sexual behaviors, including sexual revictimization, risky sexual behaviors, and teen parenthood (Trickett, Noll, & Putnam, 2011). Underlying these behavioral alterations are alterations to reproductive biology including early onset of puberty and dysregulated stress reactivity, as well as alterations to cognitive and affective functioning, such as dissociative symptoms, self-harm, and depression (Trickett et al., 2011).

Several longitudinal studies, including the Family Pathways Project, have shown that disorganized attachment behaviors in infancy are associated with elevated risk for low self-esteem, emotional instability, and internalizing and externalizing behaviors that cause relational problems with parents, peers, and romantic partners in middle childhood and adolescence (see Lyons-Ruth et al., 2008; Obsuth, Henninghausen, Brumariu, & Lyons-Ruth, 2014). Similar to the odd or contradictory presentation of disorganized behaviors in infancy, during adolescence these children show erratic behavior in the context of social interactions that is sometimes withdrawn, sometimes solicitous, sometimes controlling, and sometimes hostile. Moreover, punitive adolescent interactions were associated with poor quality romantic relationships that include abuse to and from the partner (Obsuth et al., 2014). Furthermore, research suggests that even when disorganized youth develop consistent social strategies, they struggle to adapt these strategies to different social contexts (see Cassidy, 2000). These findings suggest that disrupted attachment behaviors mediate the relationship between early adverse caregiving and lack of social-emotional competence in childhood.

Early Adversity and Attachment Disturbances in Adulthood

In accordance with an organizational perspective on development, research also suggests that individuals with aberrant or absent attachment behaviors experience difficulties in adult social and romantic relationships. There are few published reports on adult relationship outcomes in PI individuals, and many studies include individuals who entered institutional care later than the infant-toddler period. As a result, it is unclear if these outcomes are representative of adults who experienced adverse caregiving only during a discrete period of early development. Beyond this confound, the findings are mixed. For example, some research suggests that PI adults do not differ from non-adopted adults in peer problems or number of close confidents (Westhues & Cohen, 1997; Sigal, Perry, Rossignol, & Ouimet, 2003), whereas other studies suggest PI adults have fewer social supports and close confidants (Sigal et al., 2003; Weiner & Kupermintz, 2001). However, several investigators have shown that PI adults are less likely to be married than their non-adopted peers (e.g., Tieman, van der Ende, & Verhulst, 2005). Currently, to our knowledge there are no longitudinal data available on quality of adult romantic relationships for individuals who experienced early institutional caregiving. However, these data are expected to be forthcoming from the English and Romanian Adoptees Study.

The bulk of this research comes from studies using the Adult Attachment Interview (AAI; George, Kaplan, & Main, 1996). This 60 minute semi-structured interview with adults assesses caregiving experiences during childhood. Participants are asked to report general reflections on their relationship with caregivers as well as specific memories that illustrate how they feel. Additionally, adults are asked to state how these relationships may have influenced their development and current functioning. Together, these features are designed to assess adult internal working models of attachment. Adults are classified as secure/autonomous, insecure/dismissing, insecure/preoccupied, or unresolved. Research shows these classifications are stable across short-term test-re-test studies and are significantly associated with social functioning in adulthood (see Riggs, 2010). Adults who are secure/autonomous exhibit a balanced perspective of early relationships and value relationships. This classification is associated with greater capacity for intimacy, and better marital functioning and satisfaction. In contrast, adults who are insecure/dismissing deny or devalue the impact of early relationships, and often have difficulty recalling specific events. Additionally, they are more likely to both describe an early history of rejection and idealize early experiences. In relation to adult social functioning, the insecure/dismissing classification is associated with maladaptive social strategies, such as interpersonal avoidance, hostility, and holding perfectionistic expectations of social partners. Adults who are insecure/preoccupied report confusion or ambivalence about early experiences, and often describe current relationships that are characterized by active anger or passivity. This classification is associated with rumination about close relationships and interpersonal distancing. Finally, unsolved adults exhibit a mixture of dismissing and preoccupied behaviors in the context of social relationships. This classification is associated with general distrust of others, hostility, and low levels of empathy.

Studies that have empirically examined the association between infant attachment classifications and AAI classifications in samples with typically occurring variations in early caregiving experiences have found moderate to strong correspondence between secure/insecure classifications, yet suggest that experiences beyond infancy play a role in adult attachment (e.g., Waters, Merrick, Treboux, Crowell, & Albersheim, 2000). These findings provide support for an organizational perspective on development, suggesting that although social-emotional competence may initially be consolidated in infancy, there are opportunities to reorganize functioning in childhood, adolescence, and adulthood.

Finally, research that retrospectively assess childhood maltreatment and current adult social functioning show that adults who acknowledge maltreatment report elevated interpersonal sensitivity, hostility, and aggression, as well as low self-esteem, a poor or disorganized sense of self, and symptoms of dissociative disorders (e.g., see Riggs, 2010). Moreover, experiencing child maltreatment is associated with poor physical and mental health in adulthood, as well as risky sexual behavior, and unintendend pregnancy (Felitti et al., 1998). Additionally, lower quality relationships in adulthood with an individual's partner/spouse and with children (Umberson, Williams, Thomas, Liu, & Thomeer, 2015).

Intervention Research

Given strong evidence that early social interactions between infants and their caregivers provide an important foundation for subsequent development, there is a growing interest in interventions that target these interactions as a means of influencing developmental outcomes. Although correlational studies can shed light on whether measures of parent-child attachment are associated with later outcomes, this research cannot determine whether associations are causal. In contrast, intervention-oriented research can offer an effective test for causal hypotheses. Accordingly, intervention studies provide an important complement to experimental and observational work by helping clinicians and researchers discover causal mechanisms that can facilitate change in quality of infant attachment behaviors. This literature shows that early intervention can set the stage for positive developmental pathways. Intervention programs that focus on promoting attachment and relational process have also shown profound and persisting benefits on child development. Among the most well-studied of these interventions are the Bucharest Early Intervention Project (BEIP; Nelson, Fox, & Zeanah, 2014), Child-Parent Psychotherapy (CPP; Lieberman & Van Horn, 2005) the Interpersonal Therapy for Adolescents (IPT-A; Mufson, Dorta, Moreau, & Weissman, 2004), and the Skills Training in Affective and Interpersonal Regulation for Adolescents (STAIR-A; Gudiño, Leonard, & Cloitre, 2016). Many of these interventions are informed by developmental theory and guided by the organizational perspective on development.

BEIP is a preventative intervention randomized controlled trial (RCT) of a foster care intervention for children who were placed in an institution for young children in Bucharest, Romania at or shortly after birth. Children (N=136) were randomly assigned to remain in institutional care (ICG; care-as-usual group), or to be removed from institutional care and placed in a high-quality foster care group (FCG) designed specifically for BEIP. On average, FCG entered foster care at 22 months. A group of never-institutionalized children (NIG) served as community controls. Children were assessed at multiple stage-salient developmental periods, including 30, 42, 54, and 96 months. Results showed that FCG children showed improved cognitive, emotional, behavioral, and social functioning. In contrast, children who remained in institutional care showed ongoing developmental deficits across all domains. Moreover, social communication problems for ICG children increased as a function of the percentage of a child's life spent in an institution. For multiple domains relevant to social functioning (e.g., on measures of attachment security and emotional responsiveness and reactivity) FCG children performed between ICG and NIG children. Altogether, these findings demonstrate that children who experience adverse early caregiving environments can show developmental plasticity and recovery following intervention, including in domains that are relevant to long-term social functioning and relational outcomes. However, the fact that the FCG continued to perform above the ICG but below the NIG suggests that early caregiving experience can establish probabilistic parameters on development. Importantly, the inclusion of a normative community control group in the BEIP design provided a stringent test for evaluating the efficacy of the foster care intervention and the organizational perspective on development.

CPP is designed to support child development by fostering a positive, healthy parent-child attachment with at-risk children and their caregivers who have experienced some form of trauma (Lieberman & Van Horn, 2005). CPP is informed by attachment theory and the parent-child attachment relationship is conceptualized as the “client” and main recipient of therapeutic services, rather than the parent or the child independently. In addition to improving the parent-child relationship, the CPP intervention focuses on safety, affect regulation, and the joint construction of a trauma narrative. Several RCT studies have shown that CPP is effective at promoting adaptive social functioning and successful resolution of stage-salient tasks at various points in development. This includes improving attachment security in maltreated toddlers (Cicchetti, Rogosch, & Toth, 2006), and positive representations of self and caregiver in maltreated preschoolers (Toth, Maughan, Manly, Spagnola, & Cicchetti, 2002), as well as reducing emotional-behavioral problems in preschoolers exposed to domestic violence (e.g., Lieberman, Van Horn & Ghosh Ippen, 2006).

In an RCT study examining the malleability of insecure and disorganized attachment following CPP, one-year-old infants from maltreating families (N=137) and their mothers were randomly assigned to infant-parent psychotherapy (IPP), psychoeducational parenting intervention (PPI), or a community standard (CS) control group (Cicchetti et al., 2006). A fourth group of infants (N=52) and their mothers from nonmaltreating families served as an additional low-income normative comparison (NC) group. At baseline, mothers in the maltreatment group reported greater abuse and neglect in their own childhoods, more insecure relationships with their own mothers, more maladaptive parenting attitudes, and they were observed to evince lower maternal sensitivity relative to nonmaltreatment group mothers. Infants in the maltreatment groups had significantly higher rates of disorganized attachment than infants in the NC group. These findings suggest that mental representations developed from early relationship experiences may have the potential to influence parent-child interactions with the next generation. At postintervention follow-up at age 26 months, only children in the IPP and PPI groups demonstrated substantial increases in secure attachment. Moreover, disorganized attachment continued to predominate in the CS group. Interestingly, both CPP and the nonrelational intervention (PPI) were found to be equally efficacious in fostering secure attachment and in reducing disorganized attachment in infants from maltreating families (Cicchetti et al., 2006). However, when attachment was assessed again one year after the conclusion of treatment, only CPP was found to be efficacious in promoting sustained security of attachment over time (Pickreign, Stronach, Toth, Rogosch, & Cicchetti, 2013). These results suggest that, in cases of extremely maladaptive parenting, intervention models that go beyond parent skills training may be necessary.

Another RCT study of maltreated preschoolers (N=112) showed the efficacy of CPP to alter preschoolers' internal working models of attachment related to representations of themselves and their caregivers (Toth, Maughan, Manly, Spagnola, & Cicchetti, 2002). The investigators targeted these internal working models as an important outcome since they are thought to form the basis of children's future relationship expectations, as reviewed above. Importantly, results showed CPP's clinical focus on the parent-child relationship as the mechanism of change resulted in greater duration of treatment effects in behavior problems compared to standard community treatment supplemented by case management.

Findings from the Family Pathways Project, one of the few longitudinal studies prospectively examining the persisting effects of adverse early caregiving, have been important in advancing a growing understanding of how the consequences of child adversity may carry forward to influence the next generation. This 30-year study of infants of low-income mothers at-risk for maltreatment has shown that insecure and disorganized attachment is associated with maladaptive social development in childhood, adolescence, and adulthood. During the study, mother-infant dyads received weekly home visiting services. These home-visit services targeted four clinical goals: (1) building a trusting parent-child relationship; (2) increasing the family's ability to access community resources in order to meet basic needs; (3) modeling and reinforcing positive, reciprocal parent-child behavior; and (4) decreasing maternal social isolation by facilitating participation in parenting group activities. Infants (N=64) were enrolled before they turned 9 months and weekly visits continued until the child reached age 2. By 18 months of age, when attachment behaviors were measured, families had completed an average of 47 home visits. A group of nonmaltreating, low-income mothers (N=35) and infants who had never sought or received social services directed at parenting skills, or undergone extensive psychiatric treatment, provided a community comparison group. Mothers in this comparison group were individual matched to high-risk mothers based on family income, maternal education level, age, race, as well as child age, gender, and birth order. Results show that at 18 months, children in the intervention group were significantly less likely to show disorders of attachment (Lyons-Ruth, Connell, Grunebaum, & Botein, 1990). Moreover, at age five, teacher and parent ratings of child behavior problems were assessed. Results showed children who received home visits had lower teacher-rated hostile-aggressive behavior problems in kindergarten, particularly when services were delivered weekly for at least one year (Lyons-Ruth & Melnick, 2007).

IPT is an intervention for adolescents (IPT-A, Mufson, et al., 2004) and adults who have experienced early adversity and are dealing with depression and interpersonal conflict. IPT has roots in the Interpersonal Theory of Psychiatry (Sullivan, 1953), as well as attachment theory. As interpreted from an IPT framework, as well as the organizational perspective on development, relationship problems can give rise to emotional distress, and a lack of positive relationships also can have a negative effect on mental health and functioning. Informed by the organizational perspective on development, IPT-A tailors intervention efforts to facilitate the successful resolution of youth stage-salient tasks, such as establishing close peer and romantic relationships, and dealing with bereavement. Like CPP, the target of IPT intervention efforts is a primary problematic relationship. Psychoeducation and interpersonal skills-building efforts are targeted to address issues within the primary problematic relationship, with the goal of generalizing to other relational contexts. IPT theorizes that reducing relational conflict and facilitating the development of adaptive social skills will simultaneously reduce current mental health symptoms and enhance future resilience.

Although IPT is increasingly used for treatment of adolescents and adults who have experienced maltreatment (see Toth, Gravener-Davis, Guild,& Cicchetti, 2013), no study to date has explicitly adapted IPT for individuals with histories of institutional care or child maltreatment; thus, it is in need of further research attention. Existing research comparing IPT to care as usual has shown IPT to be more efficacious in reducing depressive symptoms as well as improving relationships with immediate family members for women with histories of childhood sexual abuse (Gamble et al., 2011). The provision of intervention during the adolescent period may be critical in preventing further entrenchment of maladaptive interpersonal patterns resulting from childhood maltreatment. Intervention during this period is particularly important because adolescence is a transitional turning point in biological, psychological, and social development that presents opportunities and challenges for adolescents as they embark on resolving stage-salient tasks initiating from close same-sex friends, developing an identity, and forming romantic relationships.

STAIR-A is a skill-based intervention tailored for adolescents who have experienced chronic exposure to trauma, including maltreatment, which targets emotion management and interpersonal skills (e.g., Gudiño, et al., 2016). Results of a recent pilot study shows that STAIR-A reduces depression symptoms and severity, and increases emotional and interpersonal coping competencies. Specifically, adolescent girls who participated in STAIR-A reported significantly improved perceptions of social engagement and locus of control post-treatment. This intervention approach, along with others, suggests that interpersonal functioning can operate both as a vulnerability and protective factor for adolescents.

In addition to the growing number of intervention studies, two meta-analyses provide further support that attachment-based interventions focused on promoting maternal sensitivity and/or attachment security are moderately successful (Bakermans-Kranenburg, van IJzendoorn, & Juffer, 2003 Bakermans-Kranenburg, van IJzendoorn, & Juffer, 2005). Of these, interventions focusing on maternal sensitivity and starting after the infant is six months demonstrate the best success. Interestingly, this finding is in-line with those previously discussed showing that only PI children adopted after six months old showed persistent DSED symptoms in adolescence (Rutter et al., 2010). Additionally, findings from these meta-analyses suggest that intervention studies do not need to explicitly target the reduction of frightening maternal behavior, such as negative or intrusive verbal or physical behavior (e.g., mocking or teasing infant, pulling infant by wrist), to succeed in reducing infant disorganization. Notably, Juffer and colleagues (2005) have showed that targeted changes in maternal sensitivity do not mediate the reduction of disorganized child behavior. Instead, the authors speculate that the intervention may have indirectly reduced frightening parenting behavior, perhaps by increasing parents' attention to their child's behavior as well as the impact of their own behavior on their child.

Discussion: Clinical Implications, Ongoing Debates, and Future Directions

A collective, cumulative vision is emerging regarding the processes by which childhood adversity contributes to relationship outcomes across the lifespan. First, research demonstrates that relationships with early caregivers don't just support later behavioral development, they actively shape the architecture of the brain. Accordingly, there is a growing interest in identifying the neural circuitry that support attachment and social functioning in close relationships, and characterizing neural differences associated with typical and inadequate caregiving environments. One open question this research can help answer is whether the neural circuitry that support attachment behaviors retain plasticity throughout life (Cicchetti, 2015). Although intervention studies provide some evidence of this in the form of observed changes in attachment behavior, future research is needed to provide important empirical support.

A second, associated theme is related to how early adverse caregiving environments and ongoing social difficulties may influence mental and physical health via alterations to neurobiological regulatory systems. The emerging weathering hypothesis (Geronimus, 2013) suggests that epigenetic mechanisms, which help regulate gene expression, may play a significant role in the developmental cascade initiated by early adverse caregiving. This model predicts that social adversity leads to chronic activation of stress regulatory systems that in turn increases disease vulnerability over the lifespan via epigenetic mechanisms. This hypothesis is supported by animal models (Meaney & Szyf, 2005). Additionally, recent work shows that racial discrimination, an extreme form of social adversity, is associated with a greater rate of epigenetic wear and tear on the body in young adults, and that supportive family environments measured in adolescence buffer the effect of discrimination (Brody, Miller, Yu, Beach, & Chen, 2016). Ehrlich and colleagues (2016) have shown that relationship trajectories characterized by chronic stress are associated with a chronic proinflammatory response in Caucasian adolescent girls. (Ehrclich, Miller, Rohleder, & Adam, 2016). Together, these findings emphasize the importance of healthy, supportive relationships for outcomes across development, and the need for further research examining the biological mechanisms associated with social adversity.

A third emerging theme is the continuity and discontinuity in attachment constructs. Relationships are intrinsically longitudinal and complex phenomena, calling for sophisticated statistical models and methods. As a result, the growing body of attachment work continues to motivate theoretical approaches and empirical methods that integrate interpersonal and intrapersonal analyses. For example, Fraley, Roisman, and Haltigan (2013) recently have used structural model comparisons to demonstrate that the quality of early caregiving has predictive significance that extends from infancy into adulthood in normative samples. This finding has also been replicated in a higher risk cohort (Raby, Roisman, Fraley, and Simpson, 2015), which provides strong evidence for central attachment tenants by demonstrating that caregiving shapes developmental processes that consolidate early in life and continue to influence developmental pathways over the long-term. Although there is evidence that these attachment disturbances show modest short- and long-term stability, and are risk factors for later disruptive behavior and psychopathology (van IJzendoorn et al., 1999), this modeling technique has yet to be tested with longitudinal data of children who experienced inadequate, extreme, or pathogenic early caregiving environments.

Finally, although neglect, deprivation, violence and maltreatment increase the risk of maladjustment, many children who experience these forms of adversity go on to have strong, thriving relationships. This work strongly suggests that a clearer understanding of the protective factors and processes that predict positive relationship outcomes, despite exposure to childhood adversity, has significant implications for prevention, intervention, and policy. Additionally, the clinical implications of this work are that research on resilient functioning in children who experience early adversity can shed light on the “self-righting” abilities inherent to development that clinicians and researchers can attempt to foster and facilitate (Cicchetti et al., 2015).

Conclusion

For better or for worse, early childcare experiences can have a pervasive and persistent influence on a child's development and relationships throughout the lifespan. According to attachment theory, infants are predisposed to form relationships with caregivers and learn complex social-emotional information from caregiver interactions. Research shows that early parent-child experiences shape the child's internal working models that guide the formation of future relationships. Thus, the nature of a child's early relationships has long-lasting implications for learning social competence, relationships, and development over time. Children who experience extreme or pathogenic forms of early caregiving, in the form of institutional rearing, exposure to domestic or community violence, or maltreatment, are denied opportunities for normal development normal development of social and emotional skills needed to cope with future challenges and build future social relationships. These forms of childhood adversity may set in motion a probabilistic developmental pathway, which may have detrimental effects on relationships throughout the lifespan. Research shows that these experiences are associated with negative mental representations of self and other and impaired behavioral, social, and emotional competence. In turn, these deficits can make it more difficult for children and adolescents to form peer relationships, friendships, and later adult romantic relationships. In particular, the literature examining long-term effects of inadequate and pathogenic caregiving environments have identified disorganized attachment behaviors as significant to sequelae of adverse early caregiving. Disorganized attachment behaviors stemming from adverse early caregiving and observable in infancy have been associated with persisting deficits in social competencies and social maladaptation in childhood, adolescence, and adulthood.

Additionally, early caregiving adversity disrupts a child's rapidly developing brain, with significant implications for biological and behavioral functioning (Boyce, Sokolowski, & Robinson, 2012). This theme sets the stage for questions related to remediation and recovery that are beginning to be addressed by intervention efforts. RCT studies shows that children's emotional self-regulatory and social competency skills are amenable to environmental support. For example, RAD symptoms exhibited by PI children prior to adoption decrease once they are removed from conditions of social deprivation and have an opportunity to form an attachment relationship with adoptive parents (Zeanah et al., 2015). Additionally, research has shown that CPP, an intervention that targets the dyadic relationship, decreases problematic infant behavior, even at the one-year follow-up point (Toth, Sturge-Apple, Rogosch, & Cicchetti, 2015). The fact that CPP effects strengthen and grow over time is testament to the importance of the internal working model as a developmental mechanism for carrying forward early childcare experiences across development.

Importantly, an organizational perspective on development provides a valuable framework for research considering how early adverse caregiving experiences may set up probabilistic parameters for social adaptation/maladaptation across the lifespan. By examining how stage-salient tasks at each point of development influence subsequent developmental processes and stage-salient tasks, and attending to the meaning and not merely the occurrence of a behavior, the organizational perspective on development can provide valuable insights on mechanisms that may transmit or perpetuate risk, as well as opportunities for prevention and intervention.

Relationships are foundational to the development, learning, and overall well-being of children, adolescents, and adults. Research examining relationships in at-risk populations provides an informative and complementary perspective to the examination of social relationships in a normative development (Cicchetti & Toth, 2015). By increasing the risk for profound and persisting relationship disadvantages over the lifespan, adverse early childcare experiences may have cumulative effects on mental and physical health over time. This body of work demonstrates the importance of safe, stable, and secure early relationships for children in establishing a foundation of health and well-being across the lifespan.