«Running head: Utilizing attachment measures in child custody evaluations Utilizing Attachment Measures in Child Custody Evaluations: Incremental ...»
Attachment measures and child custody evaluations
Running head: Utilizing attachment measures in child custody evaluations
Utilizing Attachment Measures in Child Custody Evaluations: Incremental Validity
Marla B. Isaacs
Department of Psychiatry
University of Pennsylvania, Philadelphia PA
Department of Psychology
Mills College, Oakland, CA
Robert S. Marvin
Mary D. Ainsworth Child-Parent Attachment Clinic
The University of Virginia
Key words: divorce, custody evaluation, adult attachment, caregiving, Strange Situation Reference citation: (2009). Journal of Child Custody, 6, 1-24.
Attachment measures and child custody evaluations 2 Abstract The American Psychological Association Guidelines for Child Custody Evaluations in Divorce Proceedings (American Psychological Association, 1994) stresses the need for a multi-method evaluation, reasoning that by considering data from different approaches, we can best ensure the validity of our findings in making custody recommendations. One of the primary issues in custody evaluations is assessing the quality of the child‟s attachment to each parent and parents‟ potential for supporting the child‟s attachment security. The case study presented in this article demonstrates how research-based standardized assessments of the child‟s attachment and parents‟ adult attachment and caregiving adds to the validity of a standardized psychological test battery typically utilized in making custody recommendations.
Attachment measures and child custody evaluations 3
Utilizing Attachment Measures in Child Custody Evaluations:
Incremental Validity The American Psychological Association Guidelines for Child Custody Evaluations in Divorce Proceedings (American Psychological Association, 1994) stresses the need for a multimethod evaluation, reasoning that by considering data from different approaches, we can best ensure the validity of our findings. In addition to individual and conjoint interviews, we can administer psychological testing and interview collateral sources (Schutz, Dickson, Lindenberger, & Ruther, 1989; Isaacs, Montalvo, & Abelsohn, 2000).
Although there is some literature (e.g., Marvin & Whelan, 2003) on the use child-parent attachment measures in custody evaluations, such tools are not typically part of the testing battery; yet, when we are evaluating custody, one of the primary issues we are assessing is the child‟s emotional attachment to each parent, and how each parent meets the child‟s attachment and exploration needs. Moreover, attachment measures are significantly correlated with important developmental outcomes for children (e.g., Sroufe, Egeland, Carlson, & Collins, 2005;
Thompson, 2008). Because of this, it is our opinion that custody evaluations can be significantly enhanced by including standardized attachment measures in the investigation. This article will demonstrate, through the use of a case study, the incremental validity that may be achieved when attachment measures are added to the battery of tests.
This is a case of a 3-year old boy, Thomas, born to unmarried parents – a 40-year-old mother, who had never been married, and a 45-year-old father, who had been married for 12 years and separated for two. Neither parent had other children. The couple met in a bookstore,
to make their relationship work, living together for a year and a half. Mother worked full time and Father worked on and off as a consultant. Thomas was cared for by an au pair. When Mother‟s company closed down, they moved together to New York City, where Mother had obtained employment and Father looked for work. Thomas was put into daycare. Father was not able to find a job, and the couple separated after two months. Thomas was now 2 years old.
Father moved nearby into his brother‟s condominium, which was on the market, and lived there for six months. During this period Mother was with Thomas 60% of the time and Father 40% of the time, alternating on a weekly basis. The parents‟ relationship was contentious. Six months later Father got a job offer in California at the same time that his brother‟s condominium sold.
Both parents petitioned the court for custody. The court kept custody arrangements the same, pending a custody evaluation. In the meantime, Father moved in to his brother‟s empty summer house, located two hours away from New York. The temporary custody order translated into 12 consecutive overnights with Mother and 9 with Father. Mother‟s work schedule required Thomas to be in daycare for 9-10 hours a day. When with Father, Thomas was in his full-time care.
Mother felt that she should have primary custody of Thomas. Father, who had been a heavy drinker, had struck her on two occasions. Following the second incident, she had threatened to leave him and take Thomas with her, but they weathered that crisis. Mother alleged that the father was an alcoholic, never had stable employment, never had a job while caring for a child and was too disorganized as a person to manage Thomas‟ caregiving. She described one time when Thomas was an infant and Father, unable to stop Thomas from crying, put the baby back in his crib in a manner that she believed was too abrupt and forceful. Mother argued that
her child, and provide for his other needs, such as setting up doctor appointments, day care, and baby play dates.
Father felt that Mother wanted him out of her life and out of the life of their son. He was remorseful that he had struck her, saying that he had never used physical force in any other relationship. He talked about the shame he felt when he put Thomas down forcefully. He felt helpless to not be able to soothe his son in the first three months of Thomas‟ life. Since his relationship with Mother ended, he explained that he has taken the role of being Thomas‟ father very seriously, and had not experienced that kind of helplessness again. He also no longer drank.
Father questioned Mother‟s mental health. He felt that she emotionally abused those close to her and was inconsistent, impatient and angry with Thomas. He worried that she cared more about what others thought about her in the role of Thomas‟ mother than about her actual relationship with her son. Furthermore, Father felt that she was hostile toward Thomas‟ relationship with him and his family, and resented his relationship with Thomas. He pointed out that he had a supportive relationship with both his family and hers, and contrasted the quality of his interactions with Mothers‟ family with what he alleged was her volatile and angry treatment of her brothers. He described Mother as having very strange ideas about the world, including being persecuted by her co-workers.
Our evaluation combined a traditional clinical assessment and an assessment of attachment and caregiving. The traditional assessment included personality testing of each parent, specifically the MMPI-2 (Butcher et. al, 2001) and the Rorschach (Exner, 2003);
conversations with collateral sources. We also reviewed court orders, prepared statements by each parent, and e-mail correspondence between them.
With regard to the assessment of attachment and caregiving, the following was
Thomas‟s patterns of interactive behavior with Father and Mother in the areas of attachment, exploration, learning, behavior management, and affect management.
Each parent‟s patterns of interactive behavior in the areas of attachment, exploration, learning, behavior management, and affect management.
Each parent‟s representations of caregiving; that is, his or her thoughts, feelings, and attitudes about being Thomas‟ parent.
Because of the well-established link between one‟s representation of one‟s own early history and parenting, representation of each parent‟s attachment experience in his or her family-of
The risk levels for psychopathology and developmental risk associated with each set of current attachment and parenting interactive and representational patterns.
The attachment-caregiving component of the evaluation is based on standardized procedures developed and validated over the past 40 years through basic psychological and clinical research worldwide. The attachment-caregiving bond is the cornerstone of the overall relationship between a child and parent (or parent-figure) (George & Solomon, 2008b; LyonsRuth & Jacobvitz, 2008; Marvin & Britner, 2008; Sroufe et al., 2005; Thompson, 2008).
Currently available cross-sectional and longitudinal studies suggest that if the attachmentcaregiving bond is ordered and within normal limits, there is a high likelihood that the child will
adaptive relationships with others and to develop a strong sense of self-reliance and social and cognitive competence. If that bond is disordered, however, there is a high likelihood that the child will experience significant current and (if disruptions in the attachment-caregiving relationship remain unresolved) future problems in emotional regulation, family relationships, teacher and peer relationships, and academic performance. Additionally, there is a significantly increased likelihood of future legal and psychiatric problems and a significantly increased likelihood that the child, when he or she reaches adulthood, will have disordered caregiving bonds with his or her own children, thus continuing the pattern into the next generation (e.g., (Hesse & Main, 2006; Solomon & George, 2006; in press).
A review of the psychological and developmental scientific literature (Cassidy & Shaver,
2008) would suggest that, after basic medical health issues, the quality of attachment-caregiving patterns may be the single most important variable in assessing and intervening in high-risk family situations, especially with younger children. These predictions of risk for various developmental outcomes reflect, of course, levels of probability, rather than certainty, for any particular child or family that is assessed.
A blind methodology was followed using three scientifically developed and validated forms of assessment – the Strange Situation (Ainsworth, Blehar, Waters, & Wall, 1978; Britner, Marvin, & Pianta, 2005; Cassidy & Marvin, 1992; Marvin & Britner, 2008; Solomon & George, 2008), the Caregiving Interview (George & Solomon, 1989; 2008a), and the Adult Attachment Projective Picture System (AAP; George & West, 2001; in press). Robert Marvin analyzed the Strange Situation; Carol George analyzed the parents‟ Caregiving Interview and AAP; and Marla Isaacs analyzed all remaining data, including the MMPI-2 and Rorschach. All analyses were
a three-way conference call. This allowed us to see to what extent our „blinded‟ findings converged or diverged, to resolve any discrepancies between the attachment and traditional clinical findings, and to determine how the assessments should be integrated into a single report.
This collaborative procedure, carefully integrating traditional and attachment measures, provides us increased scientific confidence in the overall conclusions despite the relative lack of empirical research on this population (cf., Solomon & George, 1999b). The process is also consistent with the logic of a “transportability analysis,” a form of systematic review that can allow one to generalize from results on populations that have been extensively researched to those with relatively little empirical research (see Marvin & Schutz, 2008/this issue).
Just as care is taken in traditional evaluations to construct and schedule the appointments with the child and each parent to be as similar as possible (e.g., the same time of day with each parent for young children), so too must this be done in the scheduling of the Strange Situation.
We followed the administration guidelines in the field of attachment. Care was taken such that the sessions were not administered under circumstances that attachment theorists suggest could influence parent-child interaction, such as the child being ill or just returning home to a parent following a separation. Use of the Strange Situation in a large study of infants with divorced or separated parents has demonstrated validity for the use of this procedure and session timing with this population (Solomon & George, 1999b, 1999c). Research has shown that the order and timing of administration of the adult representational measures do not influence their results (e.g., George & Solomon, 1989; George & West, in press). Validity for the Caregiving Interview has been established for divorced mothers (Solomon & George, 1999a) and fathers (Munroe, 2008). Validity for the AAP has been established in adult normative and psychiatric samples and
Psychologists can make certain inferences about how each adult might function as a parent by drawing on the results of tests of personality. These tests highlight a range of characteristics, for example, the capacity for empathy or the tendency to disordered thinking, which shape how an adult takes up the parenting role. We used the MMPI-2 and the Rorschach, instruments commonly used in a traditional evaluation, to generate our hypotheses about both parents. Table 1 presents the test results for Father and Mother.
Father Father responded openly and non-defensively to the MMPI-2 and did not elevate on any of the clinical scales. He was engaged in the Rorschach, thus providing us valid information from two independent tests.
Father has narcissistic vulnerabilities. He may feel easily slighted and defensive when others express doubts about his decisions or behaviors, though he is unlikely to reveal such feelings to others. A highly introspective person, he prefers to process those feelings internally. He uses intellectualization to distance himself from strong emotions, thus muting the intensity of his feelings. But under the sway of intense criticism, he is likely to erupt, as when he struck Mother. Given that his psychological resources are considerable, and his defenses strong,
Though flexible in his thinking, Father does not always think clearly and coherently.