Brief History of the Research and Development of COS Interventions

Glen Cooper, Kent Hoffman and Bert Powell are the founders of Circle of Security International. They have worked together in clinical practice since 1985, and had been friends and colleagues for a decade before that.  Their shared clinical interest began with their work in family therapy and systems theory, which, after many years, led them to object relations theory.  It was through their study of object relations theory that they were introduced to attachment theory.

In 1993, they met Jude Cassidy in Seattle while attending her training on the preschool attachment coding system.  They convinced her to teach them her graduate attachment seminar via weekly telephone calls.  What began as a one-semester course evolved into two years of fruitful study and an ongoing working relationship and friendship with Dr. Cassidy.  After several more years of study, Glen, Kent, and Bert were excited to begin the process of creating a clinical application that combined family systems, object relations, and attachment theory.

The Circle of Security began in 1998.  After consulting with Spokane Head Start for years, Kent, Glen, and Bert began to apply attachment theory to the Early Head Start home visiting program. They also piloted a group model with several groups of homeless/street dependent parents. The director’s administrative assistant came across a Head Start University Partnership Grant application that was relevant to that work.  She went to the director, Patt Earley and said, “I have two questions, isn’t this what we are doing and why are we not being paid for it?”

In 1998, Bert, Kent, and Glen were asked by the director of Spokane Head Start to apply for a University-Head Start Partnership grant.  They invited Bob Marvin to be the principle investigator and received a three-year grant.  The results were published in the Journal of Consulting and Clinical Psychology (JCCP) in 2006.

Through the USDHHS grant, the original 20-week clinical group protocol was developed.  Parents were videotaped for a pre-intervention Strange Situation Procedure (SSP).  That video footage was  coded for research purposes, and clinicians edited clips to be shown to the parents during the group.  The SSP made it possible to discern parents’ interactional struggles with their children based on attachment theory.  In addition, a parent perception interview, the Circle of Security Interview (COSI), made it possible for clinicians to determine a parent’s defensive structure based on object relations theory. Using this information, a series of video clips from the SSP were specifically selected for each parent to be shown during the group.

The results of this intervention showed that there was a significant decrease in disorganized attachment status from pre- to post-intervention; at baseline, 60% of children were categorized as disorganized, while at follow-up only 25% were so categorized. Also, there was a significant decrease in insecure attachment status from pre- to post-intervention. At baseline, 80% of children were categorized as insecure, while at follow-up only 46% were so categorized.

Hoffman, K., Marvin, R., Cooper, G. & Powell, B. (2006). Changing toddlers' and preschoolers' attachment classifications: The Circle of Security Intervention.Journal of Consulting and Clinical Psychology, 74, 1017-1026.

You can read the JCCP article here: JCCP COS Published Article.

In 2002, in conjunction with Jude Cassidy, Cooper, Hoffman, and Powell developed a four session home visitation protocol (COS-HV4) for economically stressed mothers with irritable infants.  The study’s goal was to examine the moderating effects of infant irritability and maternal attachment on the effectiveness of the COS-HV4 at reducing the rates of insecure infant-mother attachment.  The four-session COS-HV4 intervention took place during three home visits lasting an hour and a half each, and one brief fourth follow-up visit between infant ages 6.5 and 9 months.

Results indicated that for dyads that were particularly at-risk for insecure infant attachment (e.g., a dismissing mother with a highly irritable infant) the intervention significantly reduced the risk of insecure attachment.

Cassidy, J., Woodhouse, S., Sherman, L., Stupica, B., & Lejuez, C. (2011). Enhancing infant attachment security: An examination of treatment efficacy and differential susceptibility. Journal of Development and Psychopathology, 23, 131-148.

In 2003, again in conjunction with Jude Cassidy, Cooper, Hoffman, and Powell developed a COS psychoeducational program for Tamar's Children (jail diversion program; women start in their last trimester and stay in the program until their child is one year of age).  The results of this COS intervention were published in a special issue of Attachment and Human Development in 2010. Twenty mothers completed treatment and were seen with their 12-month-olds in the Ainsworth Strange Situation attachment assessment.

Fourteen of the 20 infants (70%) were classified as securely attached to mother.  Though, clearly, pre-intervention Strange Situation assessments could not be done prenatally, this rate of security done when the child reached 12 months, is significantly higher than rates typically observed in samples of high-risk mothers, and was identical to the rates typical of low-risk, middle-class samples. In addition, only four infants (20%) were classified as insecure/disorganized, the insecure subgroup with highest risk for psychopathology. This rate of disorganization is significantly lower than that found in at-risk samples, and it is identical to the rate that is typical of low-risk, middle-class samples.

Cassidy, J., Ziv, Y., Stupica, B., Sherman, L. J., Butler, H., Karfgin, A., Cooper, G., Hoffman, K. T., & Powell, B. (2010). Enhancing maternal sensitivity and attachment security in the infants of women in a jail-diversion program. In J. Cassidy, J. Poehlmann, & P. R. Shaver (Eds.), Incarcerated individuals and their children viewed from the perspective of attachment theory. Special issue of Attachment and Human Development.

The original Circle of Security clinical model, although shown to be effective, is time and resource intensive. Encouraged by the results of the 4 week home visitation model and in response to the call for a more scalable intervention, in 2007 Cooper, Hoffman, and Powell started development of an eight-session parent reflection protocol, Circle of Security Parenting (COS-P).  The idea was to create COS-P on the same theoretical foundation as the COS clinical approach but rather than using individualized video reviews, facilitators would use the same DVD footage.

Video clips of parents and children were developed to introduce parents to attachment theory in an accessible manner. Parents are provided an opportunity to enhance their observational and inferential skills, and invited to engage in reflective dialogue regarding their strengths and struggles in parenting. Over the course of eight sessions, the focus of the intervention moves from discussing secure attachment and children's needs, to the more vulnerable process of parents reflecting on themselves and the defensive behaviors that maintain insecure and disorganized attachment.

During the three years of production, the DVD was piloted in six countries at 30 different sites and received very positive feedback and suggestions, which were incorporated into the final version.  In 2010 the COS-P DVD was released and has met with tremendous success.  Over 15,000 people have been trained in this model and demand for translations has been high. To date, the COS-P DVD has been translated into: Spanish, Italian, Japanese, Danish, Norwegian, Swedish, Romanian, Mandarin and Australian/English.  Soon we will be able to add French and Cantonese to this list. Preliminary research on this scalable version of COS is very promising and RCTs are currently being conducted in the U.S., Australia, and New Zealand.