Supporting Families to Build Secure Attachment Relationships : Comments on Benoit, Dozier, and Egeland

Centre for Child & Family Studies, Leiden University, Netherlands
, 2nd ed.

PDF version


Since Bowlby and Ainsworth formulated attachment theory,1,2 many early intervention programs have been launched that aim to promote secure child-parent attachment relationships. Usually, these intervention programs are designed to enhance parental sensitivity, the ability to accurately perceive children’s attachment signals, and the ability to respond to these signals in a prompt and appropriate manner.2 The ultimate goal of these interventions is to turn insecure-avoidant (A) and insecure-resistant (C) attachment relationships into secure (B) child-parent attachment relationships.2 In a few programs, the intervention is not only directed at sensitive parental behaviour but also at maternal mental attachment representations, as in the STEEP (Steps Toward Effective Enjoyable Parenting) program described by Egeland. According to Benoit, with the discovery of a new insecure attachment category, disorganized attachment (D),3 new challenges arose for attachment-based interventions. Because of the negative impact of, in particular, disorganized attachment on child outcomes, attachment-based interventions should not, or not only, focus on the empirically derived determinants of organized (A, B, and C) attachment, such as parental (in)secure mental attachment representations and sensitive behaviour (see Dozier), but also on the determinants of disorganized (D) attachment. Empirical studies have found evidence for Main and Hesse’s4 model that parents’ unresolved loss or trauma is linked to children’s insecure-disorganized attachment through frightening or frightened parental behaviour. However, there are as yet no reported outcomes from interventions that have directly targeted frightening behaviours. As a first step, it is important to evaluate the effects of attachment-based interventions that include infant attachment disorganization as an outcome measure (see below), but in the next step interventions that are specifically designed to prevent insecure disorganized attachment should be tested.

Research and Conclusions

Egeland elegantly summarizes the main tenets of attachment theory. According to Bowlby,1 infants are biologically predisposed to use their parent as a haven of safety to provide comfort and protection when they are distressed, and as a secure base from which they can explore the environment. As children develop, they form mental representations or inner working models on the basis of their experiences with their caregivers. If children have had positive experiences with sensitive parents, they will continue to rely on them by showing their distress and being calmed by contact with the parent (defined by Ainsworth2 as secure patterns of attachment). In contrast, insensitive parents reject their children’s bids for comfort, and other parents are inconsistently available. Children of these parents develop insecure attachment relationships, either avoiding, or angrily or passively resisting the parent. Secure attachments during early childhood predict more optimal developmental outcomes in later childhood (e.g. social competence), whereas insecure attachments predict less optimal child outcomes. Drawing on the many positive outcomes of secure attachment found in empirical studies, Egeland comes to a crystal-clear conclusion that programs should be designed and evaluated to promote secure attachment relationships in order to improve developmental outcomes of children who are at risk for poor developmental outcomes. Egeland reviews several attachment-based interventions (e.g. the comprehensive STEEP project). As well, a first meta-analysis in this field5 is described. This meta-analysis of the effects of 12 attachment-based interventions on maternal sensitivity and infant security showed that these interventions were more effective in changing parental insensitivity than in changing children’s attachment security.5

Egeland does not address the follow-up of this first meta-analysis on parental sensitivity and attachment, nor does he cover the question of how insecure disorganized attachments might be prevented. Recently, 88 interventions on maternal sensitivity and infant security in 70 studies were included in a thoroughly extended and updated quantitative meta-analysis.6 This meta-analysis showed that interventions that specifically focused on promoting sensitive parental behaviour appeared to be rather effective in changing insensitive parenting as well as infant attachment insecurity. One of the conclusions of this series of meta-analyses, also illustrated in the title of the paper “Less is more,” was that interventions with a modest number of intervention sessions (up to 16) appeared to be more effective than interventions with larger numbers of sessions, and this was true for clinical as well as for non-clinical groups.6 This diverges from Egeland’s conclusion that more comprehensive, long-term interventions are necessary for high-risk families. Although this might be true for other intervention goals, such as helping high-risk mothers to cope with adversity or the daily hassles surrounding the birth of a child, the recent meta-analysis shows that for sensitivity and attachment, the most effective way is to provide attachment-based interventions in a modest number of sensitivity-focused sessions.

Dozier elaborates on parental state of mind as one of the strongest predictors of infant attachment. Parents who are able to reflect on their own childhood experiences in a coherent way are said to have autonomous states of mind. When parents are not coherent in discussing their own attachment experiences, they are said to have non-autonomous states of mind. Here, the work of Main comes to the fore: the Adult Attachment Interview7 enables coders to distinguish reliably between parents with insecure (dismissing, preoccupied or unresolved) states of mind and parents with secure (autonomous) attachment representations. Several empirical studies and a meta-analysis8 have found that insecure parents usually have insecurely attached infants and secure parents tend to have secure children. Dozier remarks that some attachment-based interventions are designed to target parent state of mind as a means of changing infant attachment, although many other interventions try to change parental sensitivity alone.

Citing the recent meta-analysis of attachment-based interventions by Bakermans-Kranenburg and colleagues,6 Dozier summarizes the main outcomes: brief sensitivity-focused interventions that start after the child is at least six months old are most successful, irrespective of parental risk status or socioeconomic status. Dozier does not explicitly address disorganized attachment and the implications of disorganized attachment for intervention research.

In contrast to the first two authors, Benoitexplicitly describes the challenge of the discovery of insecure-disorganized attachment for the field of attachment-based interventions. At the beginning of her paper, she notices that of the four patterns of infant attachment (secure, avoidant, resistant, disorganized), the disorganized classification has been identified as a powerful childhood risk for later psychopathology. She continues with the observation that for disorganized attachment the focus of the intervention should not be parental sensitivity, as she notes that sensitivity is not linked to disorganized attachment. Nevertheless, a meta-analysis showed that interventions with a focus on sensitivity were successful in reducing or preventing attachment disorganization9 (see below), and we noted that the explanation for this finding might be that parents become more focused in the interaction with their child, and thereby less prone to dissociative processes in the presence of the child. According to Benoit, one recently identified pathway to disorganized attachment is children’s exposure to specific forms of aberrant caregiving behaviours that are referred to as “atypical.” Therefore, Benoit concludes that attachment-based interventions should focus both on improving parental sensitivity (to promote secure attachment) and on reducing or eliminating atypical parental behaviours (to prevent or reduce disorganized attachment). Benoit’s own study, which demonstrated the effects of a brief, focused, behavioural parent training intervention in reducing atypical caregiver behaviours, is a first example of much needed studies designed to reduce frightening/frightened or atypical parental behaviours. It would be exciting to learn whether this type of intervention was indeed successful in preventing or reducing disorganized attachment.

Implications for Clinical Practice and Services

What can we conclude about attachment-based interventions and the state of the art of intervention research? Based on the two meta-analyses5,6 conducted in 1995 and 2003, several conclusions for clinical practice and services can be drawn. It has been empirically proven that interventions can successfully enhance parental sensitivity and promote secure attachment in children, in particular when the intervention is relatively brief (up to 16 sessions), behaviourally oriented, focuses on sensitivity only (instead of broader interventions including social support, etc.), and starts after the infant’s age of six months. However, long-term and broadly-focused support of multi-problem families in coping with their daily hassles may be needed in order to enable them to focus on sensitivity subsequently.6 The 2003 meta-analysis also found an important dose-response relation between the success of the intervention on parental sensitivity and its impact on children’s attachment security: only interventions that brought about substantial effects on sensitivity succeeded in changing attachment insecurity.6

Both meta-analyses included interventions designed to change children’s insecure, organized attachment relationships: insecure-avoidant and insecure-resistant relationships, and not the clinically important category of insecure-disorganized attachment. Today, few interventions have been specifically designed to prevent attachment disorganization. In the same vein, most attachment-based interventions do not report effects on disorganized attachment. This is a serious gap in our knowledge for two reasons: (1) Recent research has shown that disorganized attachment is a predictor of psychopathology, whereas insecure-avoidant and resistant attachment lead to less optimal but not pathological child adjustment.10 Therefore, it is imperative to evaluate attachment-based interventions on their potential value to prevent attachment disorganization. (2) Because even secure children are considered insecure when their attachment behaviour shows serious signs of disorganization, it is of great relevance for interventions to report not only effects on secure attachment but also effects on disorganized attachment.

Recently, a narrative review and quantitative meta-analysis has been completed including 15 preventive interventions that included infant disorganized attachment as an outcome measure.9 Although the overall effect of all interventions combined was not significant, some interventions did succeed in preventing disorganized attachment in children. These interventions shared the following characteristics: They started after six months of the infant’s age rather than before six months; they were sensitivity-focused; and they involved samples with children at risk rather than at-risk parents.9

As an example, a preventive intervention in families with internationally adopted infants significantly enhanced maternal sensitivity and also significantly reduced disorganized attachment: in the intervention group there were only 6% disorganized-attached children compared with 22% in the control group.11 This study used a brief intervention of three home-based sessions of video feedback focusing on parental sensitivity, with the intervention starting when the child was six months old. Based on the positive outcomes of this study, adoption practice in the Netherlands has changed. New adoptive parents can apply for a new adoption after-care service: up to four sessions of video feedback, implemented by a central adoption service organization financed by the government. An increasing number of adoptive parents make use of this new service.  The video-feedback intervention used in adoptive families11 was extended and adapted into the Leiden VIPP (Video-feedback Intervention to Promote Positive Parenting).12, 13 The VIPP program and several adaptations and extensions have been used in different cultures and contexts, for example with insecure or eating-disordered mothers, in families with premature and sick infants or externalizing toddlers, and in a daycare setting.14

Future studies should also focus on evaluating interventions that are explicitly directed at parental frightening or frightened behaviour as the empirically derived determinant of infant disorganized attachment. As the meta-analyses on organized and disorganized attachment all indicate an important role for parental sensitivity, it may be wise to include the enhancement of parental sensitivity in all attachment-based interventions. 


  1. Bowlby J. Attachment. New York, NY: Basic Books; 1982. Attachment and loss. 2nd ed; vol 1.
  2. Ainsworth MDS, Blehar MC, Waters E, Wall S. Patterns of attachment: a psychological study of the strange situation. Hillsdale, NJ: Lawrence Erlbaum Associates; 1978.
  3. Main M, Solomon J. Procedures for identifying infants as disorganized/disoriented during the Ainsworth Strange Situation. In: Greenberg MT, Cicchetti D, Cummings EM, eds. Attachment in the preschool years: Theory, research, and intervention. Chicago, Ill: University of Chicago Press; 1990:121-160.
  4. Main M, Hesse E. Parents’ unresolved traumatic experiences are related to infant disorganized attachment status: Is frightened and/or frightening parental behavior the linking mechanism? In: Greenberg MT, Cichetti D, Cummings E, eds. Attachment in the preschool years: Theory, research, and intervention. Chicago, Ill: University of Chicago Press; 1990:161-182.
  5. Van IJzendoorn MH, Juffer F, Duyvesteyn MGC. Breaking the intergenerational cycle of insecure attachment: A review of the effects of attachment-based interventions on maternal sensitivity and infant security. Journal of Child Psychology and Psychiatry 1995;36(2):225-248. 
  6. Bakermans-Kranenburg MJ, Van IJzendoorn MH, Juffer F. Less is more: Meta-analyses of sensitivity and attachment interventions in early childhood. Psychological Bulletin 2003;129(2):195-215.
  7. Main M, Goldwyn R. Adult attachment rating and classification system.Berkeley, Calif: University of California. Unpublished manuscript.
  8. Van IJzendoorn MH. Adult attachment representations, parental responsiveness, and infant attachment: A meta-analysis on the predictive validity of the Adult Attachment Interview. Psychological Bulletin 1995;117(3):387-403.
  9. Bakermans-Kranenburg MJ, Van IJzendoorn MH, Juffer F. Disorganized infant attachment and preventive interventions: A review and meta-analysis. Infant Mental Health Journal. In press.
  10. Solomon J, George C. The place of disorganization in attachment theory: Linking classic observations with contemporary findings. In: Solomon J, George C, eds. Attachment disorganization. New York, NY: Guilford Press; 1999:3-32.
  11. Juffer F, Bakermans-Kranenburg MJ, Van IJzendoorn MH. The importance of parenting in the development of disorganized attachment: evidence from a preventive intervention study in adoptive families. Journal of Child Psychology and Psychiatry 2005;46(3):263-274.
  12. Juffer F, Bakermans-Kranenburg MJ, Van IJzendoorn MH. Enhancing children’s socioemotional development: A review of intervention studies. In: Teti DM, ed. Handbook of Research Methods in Developmental Science. Oxford, United Kingdom: Blackwell Publishers; 2004:213-232.
  13. Juffer F, Bakermans-Kranenburg MJ, Van IJzendoorn MH. Introduction and outline of the VIPP and VIPP-R program. In: Juffer F, Bakermans-Kranenburg MJ, van IJzendoorn MH, eds. Attachment-based intervention with video-feedback and biographical discussion: The Leiden VIPP and VIPP-R program. Mahwah, NJ: Lawrence Erlbaum. In press.
  14. Juffer F, Bakermans-Kranenburg MJ, Van IJzendoorn MH, eds. Attachment-based intervention with video-feedback and biographical discussion: The Leiden VIPP and VIPP-R program. Mahwah, NJ: Lawrence Erlbaum. In press.

How to cite this article:

Juffer F, Bakermans-Kranenburg MJ, van IJzendoorn MH. Supporting Families to Build Secure Attachment Relationships : Comments on Benoit, Dozier, and Egeland. In: Tremblay RE, Boivin M, Peters RDeV, eds. van IJzendoorn MH, topic ed. Encyclopedia on Early Childhood Development [online]. Updated: March 2007. Accessed June 14, 2024.

Text copied to the clipboard ✓