Enhancing Early Attachments: Theory, Research, Intervention, and Policy

This edited book is part of a series aimed at bringing research on child development to bear on public policy. It originated in a conference held at Duke University in the fall of 2003 on the topic of interventions to address dysfunctions in the parent-child attachment relationship.

The book is divided into three sections. The first section addresses the theoretical and empirical bases for interventions aimed at promoting improvements in attachment relationships between mothers and young children. The second section highlights various programs some of which focus exclusively on the attachment relationship while others include an attachment focus as part of a broader intervention. The third section offers commentaries by known luminaries in the field with discussions about current research questions, such as short term focused versus more broadly focused programs, and implications for public policy.

Lisa Berlin begins with a critical review of the current state of research on early intervention. She discusses, first, the prevailing model of transmission of attachment security that posits that parent’s state of mind (internal working model) contributes to parental behaviour (sensitivity) and this to parent-child attachment security. The fact that research has shown only modest correlations between parental sensitivity and parent-child attachment, generally less than that between parent’s internal working models and parent-child attachment, is referred to as the “transmission gap”. This suggests that sensitivity per se may not fully mediate the transmission process. Possible explanations for this discrepancy include lack of sensitivity of measurement approaches but also lack of attention to contextual factors that could be expected to affect various aspects of the model, such as stress and parental mental illness.

She calls for further research that examines a broader array of factors that may impact on intervention. Berlin then proceeds to outline three theory- and research-based therapeutic tasks expected to contribute to a more refined transmission model. These include more attention to enhancing parental reflective function with respect to their own attachment experience and how that may impact their relationship with their infant, promoting parenting behaviours that more directly reflect Bowlby’s notion of the parent as a “secure base” from which the child explores and lastly, examining how the parent-therapist relationship acts as a “secure base” for change in the parent.

Berlin proceeds to re-examine the present controversy in early intervention research stemming from the recent and conflicting meta-analyses that suggest that “less is more” (short, focused interventions are more efficacious) and that “more is better” (longer term, insight oriented interventions are better). These reviews based on roughly the same studies have provoked vigorous debate as to the appropriate targets and duration of interventions. She concludes that, as the effects of interventions are relatively weak, when measured rigorously, there is no clear cut answer to this debate but research should attempt to address aspects of the transmission model looking more systematically at factors that may make shorter interventions work for some while longer interventions may be needed for others. This includes a more systematic examination of the impact of interventions on those factors assumed to mediate the process. This chapter is rich and worth reading several times to appreciate its wealth of ideas.

In Chapter Two, again in an attempt to understand the “transmission gap” phenomenon, Cassidy and collaborators describe their preliminary findings from The First Year Project, an intensive exploration of maternal behaviours that seem related to infant attachment. They conclude that the global concept of maternal sensitivity, as presently measured, may not detect variations in the way that sensitive and insensitive mothers may be able to provide a secure base when the attachment system is activated. They speculate that even insensitive mothers may respond eventually to an infant’s efforts at obtaining mother’s soothing. They do, however, comment that some behaviours, such as, frightening reactions, overt rejection and hostility, always seem to predict to an insecure attachment. They encourage research aimed at clarifying some of their speculations.

Yair Ziv provides a brief overview of how attachment interventions may add to, or elucidate, attachment theory. In a chapter entitled The Developmental Neurobiology of Disrupted Attachment: Lessons from Animal Models and Child Abuse Research Frank Putnam reviews the animal work that leads to alterations in stress reactivity. He discusses Meaney’s work with high and low “licking and grooming behaviour” (LG) rat mothers as well as the separation studies that can increase resilience (high LG) and alternatively increase stress reactivity (low LG and long separations). Suomi’s studies with rhesus macaques using early separation stress and cross-fostering, along with genetically predisposed stress reactive offspring, are used to illustrate the interaction between genes and the environment. Putnam then reviews the studies on maltreatment and their impact on the hypothalamic axis. He points out that traumatized individuals may demonstrate hypo- as well as hyper-cortisolemia, a situation that still lacks a compelling theory. Perhaps, more importantly the evidence from both animal and human studies suggests that traumatized individuals generally display long term alterations in stress reactivity. Putnam’s chapter is a good overview of the neurobiology of stress reactivity and its connections with later disorders such as depression. It is less clear how these findings relate to attachment interventions although the Meaney work on high LG mothers is suggestive of the importance of handling or close physical contact. It is somewhat surprising that no mention is made of the Prairie and Montane voles, animals that differ markedly in their social interaction. Although the overall levels of brain oxytocin and vasopressin (thought to be involved in social behaviour) do not differ in these two species, there is a difference in the areas of the brain in which they are highest. In the Prairie voles (social species) oxytocin and vasopressin are found in greater amounts in the centres of the brain connected with reward. Comparisons have been made with very social/altruistic individuals where studies have demonstrated activation of similar circuits in a social co-operation task. (See Charney 2004 for an excellent review of neurobiology of stress reactivity). This work is suggestive of variations in maternal and child neurobiology that may predispose to greater or lesser capacity in making attachments.

Alicia Lieberman and Lisa Amaya-Jackson offer a compelling argument from a dynamic perspective that many children labeled as having a disorganized attachment may be displaying symptoms of PTSD. Through a case illustration they show how the child’s reaction to trauma when his mother may also be traumatized may produce behaviours that resemble a disorganized attachment. They argue that being aware of a history of traumatization may enhance understanding of the presentation and lead to a reworking of the trauma that decreases the child’s sense of helplessness, thus contributing to increased security in the parent-child relationship. I think this kind of examination of the notion of disorganized attachment is very important. Beginning to look at other aspects of both a child’s experience as well as his genetic predisposition to dysregulation is likely to provide a more clear understanding of this important area.

Beginning the section on interventions, Cooper, Hoffman, Powell and Marvin provide an overview of their 20 week group intervention called The Circle of Security Project. This program blends attachment with object relations theory. They teach parents about the child’s needs in developing a secure attachment, that is, the need to explore versus the need to return to a secure base. They, then help parents learn to identify behaviours in both parent and child that miscue the other and, although designed to reduce discomfort, lead to skewed attachment patterns. Understanding why a parent or child reacts defensively is part of the learning about how past events affect present behaviour. They distinguish their intervention from a parent-education model by its emphasis on enhancing relationship capacity through identifying and exploring defensive processes that interfere with relationships. This is an interesting and well described intervention that is intuitively appealing. Evaluation of its efficacy is needed.

Arietta Slade and collaborators describe the theory behind their program “Keeping the Baby in Mind”. This dynamically oriented approach developed out of the seminal work by Fonagy, Target and others on reflective functioning (RF). These authors have speculated that a parent’s capacity to understand their own and their child’s feelings and desires enables a parent to respond sensitively to their child and thus promote a secure attachment. Following on this thinking and combined with their own research that showed a high correlation between maternal RF and adult and child attachment, Slade et al designed a program to enhance maternal RF. This program, aimed at high risk families, uses both Masters level nurse and social workers in a home visiting program. There is a dual emphasis on physical and mental health support. Although currently there is only pilot data, a randomized control trial is underway. Anecdotal evidence from the pilot study is highly suggestive that maternal reflective functioning can be enhanced and that doing so impacts positively on parent-child attachment and other developmental outcomes.

Dealing with the complicated needs of children in foster care motivated Dozier and colleagues to develop the Attachment and Biobehavioral Catch-up (ABC) Program. This innovative 10 week program operates on four components felt to be particularly germane to the attachment difficulties typical of infants in care. Recognizing that foster parents with dismissing or unresolved states of mind are likely to have children classified as disorganized, they support parents to be optimally nurturant especially when this does not come naturally for them. Secondly knowing that many children placed in care often give messages that they do not need or want care, foster parents are encouraged to provide nurturing care even when infants appear not to need or want that care. Common regulatory problems such as eating and sleeping are approached through touch and following the child’s lead. Lastly knowing that many children in care are very sensitive to threat cues, the program emphasizes the importance of diminishing threatening behaviours in foster parents. Case studies suggest the likelihood of increasing secure attachment in these children but the efficacy of this intervention must await the current RCT for definitive results.

Zeannah and Smyke describe two programs developed to deal with children exposed to severe neglect. The first is a clinical intervention for maltreated children removed from birth families in New Orleans and the other a research study comparing children in foster care versus institutional care in Romania. The authors point out the gap between clinical and developmental approaches to understanding attachment, the clinical providing diagnoses such as Reactive Attachment Disorder versus the developmental approach emphasizing security versus insecurity. Their approach builds on strong clinical skills in supporting nurturant parenting and reading the child’s cues but emphasizes the parallel process of relationship security between therapist and foster mother creating a similar relationship between foster mother and child. Anecdotal evidence suggests improvement in attachment security in foster mothers and their children but less impact on decreasing the indiscriminate friendliness of many of these children.

David Olds provides an updated overview of the Nurse-Family Partnership with long term results from several of the research sites. The impact of this very important intervention is most apparent with high risk families and has allowed Olds to argue that it is not logically delivered as a universal program because of its cost and limited evidence of benefit in middle class families. The attachment component is emphasized in the teaching of sensitive responsive parenting behaviour but also in the relationship between nurse and mother. Olds is careful to point out that in high risk families much attention needs to be given to a wide range of family needs, including but not limited to sensitive parenting. Further he argues the importance of fidelity to the model particularly the use of highly trained professionals such as nurses.

In a chapter on Enhancing Early Attachments in the Context of Early Head Start, Spieker and colleagues review their rather disappointing results from a Parent-Child Communication Coaching Program, added on to an Early Head Start program. Despite being theory-based and ecologically-grounded they had trouble retaining subjects and concluded that what they learned most was the multitude of real-world problems that had to be navigated in transferring research into effective programs.

As the last chapter in the section on Interventions Nagle and Wightkin describe the onerous and complicated process that preceded the implementation of The Nurse-Family Partnership in Louisiana. Although not highly relevant to the Canadian scene it does provide an overview of the multitude of players who need to be brought onboard in order to leverage support for broad early intervention initiatives.

Beginning the commentaries, IJzendoorn, Bakermans-Kranenburg and Juffer review their various meta-analyses on changing parent sensitivity and infant attachment security. Their conclusion “Less is More” is supported by careful work and argued well. They advocate for brief, behaviourally focused interventions that attempt to increase parental sensitivity, and through this, enhance parent-child security. They do acknowledge that more studies are needed that focus on preventing disorganized attachment through decreasing frightening maternal behaviour. This is a good overview of one side of the argument.

O’Connor and Nilsen comment on the confusion in the minds of many about the difference between attachment security and attachment disorders. They argue for more attention to attachment disorders and more information about attachment to prevent the use of therapies such as “holding therapies” that actually contravene attachment principles. This is further illustrated in their comments about the needs of foster parents when attempting to address the behavioural and emotional needs of maltreated foster children.

In the final commentary Greenberg summarizes achievements in attachment research and points to the need for further well designed studies that can begin to answer ‘what works for whom’. He emphasizes the need to examine differences between populations who may be “at risk” versus those who demonstrate clinical problems with attachment. This involves more careful assessment prior to intervention to ascertain a variety of risk factors that may be relevant to outcomes. He also raises the issue of what is the appropriate outcome, change in attachment security versus changes in other aspects of parenting behaviour that may relate to later social and emotional development.

This book is a valuable resource for anyone working in the area of early intervention. Not all chapters are equally useful, but on the whole it provides a useful update in this rapidly advancing field.