Mary Main, Professor of Psychology at the University of California, Berkeley, has received a Fund for Psychoanalytic Research grant in support of her work on processes of attachment. Her research program, begun over two decades ago, promises to be of great interest to psychoanalysts. Although analysts have in the past been critical of attachment theory, it is an area of inquiry currently undergoing an interesting methodological transformation.
A great deal of research, interesting and important in its own domain, has limited applicability to the concerns of psychoanalysis. Certain methods, however, yield data closer to these concerns than do others. Most standard between-group designs, for instance, which compare experimental and control groups, are not meant to yield a dynamic profile of an individual in the throes of personality change over time, and this after all is the clinical data base native to psychoanalysis. Similarly, many other conventional parametric statistical procedures engender a static image in which crucial interindividual variability is washed out and treated as “error.” An alternative is longitudinal research, but even here one must distinguish between retrospective and prospective designs. In retrospect, things tend to appear clear and inevitable, a situation yielding illusory pictures of correlation that are easy to confuse for causal relationships. By contrast, although in the prospective view all one sees is uncertainty, causes that can be inferred as work proceeds tend to be less specious. Consequently, the Fund for Psychoanalytic Research looks to sponsor projects with a prospective method and data base. This implies sequential sampling of data, the following of individuals over time, and linking early time points, or process variables, to outcome criteria. This approach characterizes several burgeoning areas of research, as methodological innovations are being developed in such areas as psychotherapy research, lifespan human development, and econometrics.
Within this genre of research, Main proposes to study adolescents over many years, a method that should be particularly appealing to psychoanalysts, who likewise study lives over time. To aid her in this work, Main has developed the Adult Attachment Interview (AAI), an instrument designed to provide access to and provide a classificatory framework for an “adult's state of mind with respect to attachment.” It consists of a set series of 15 questions that call for both descriptions and evaluations of important childhood experiences and how these may have affected the person's development. The subject thus produces and reflects upon crucial memories while trying to maintain coherent discourse with the interviewer. The interview is then transcribed verbatim and scored with respect to a person's current state of mind pertaining to early experiences and their effects. A crucial aspect of the scoring involves the subject's ability to maintain coherent discourse while describing the memories. This measure ignores the stated content of the experience and instead uses “discourse analysis” to identify variables from which behavior may be predicted. This analysis allows the classification of discourse into two major categories: secure/autonomous and insecure/incoherent. Patterns of discourse falling within the latter category can in turn be divided into subcategories of insecurity with respect to early experience: dismissing, preoccupied, and unresolved/disorganized. Each of the interview categories was in earlier research found to be linked to one of the categories of infant response to the parent in the Ainsworth Strange Situation, a laboratory-based separation and reunion sequence that is an observational procedure yielding three types of attachment. Main's AAI instrument has proven psychometrically sound, reliable across judges and interviewers, and unrelated to intelligence or verbal fluency.
In the work underwritten by the Fund, Main and her collaborators in Cermany were given access to 50 adolescents whose relationship with their parents had been assessed using the Ainsworth Strange Situation at the age of 12 months and again at 6 years. Beginning in 1977, Main and her Berkeley colleagues had similarly assessed 70 now-adolescent subjects at the same ages. These Berkeley adolescents will be available for reassessment over the next three years. This proposed cross-national study will attempt to determine whether AAI status at age 17 is predictable from early attachment experiences with parents, as assessed using the Ainsworth Strange Situation.
Main's proposal also addresses a recent line of research on attachment processes, developed by Giovanni Liotti in Rome, that attempts to determine whether individuals classified in infancy as disorganized/disoriented with their mother will be more likely than others to respond with dissociation to subsequent trauma. Main will investigate this question using a measure of dissociative experience with these adolescents.
A number of novel hypotheses will be explored. For instance, Main will investigate whether AAI status at age 17 is predictable from early attachment status. The question will also be explored of whether a person's state of mind with respect to a remembered childhood experience follows from reality elements of that experience.
Findings using the AAI that are significant for psychoanalysis have already emerged. It is not what a parent has actually experienced in the past but rather how experiences are currently related to an interviewer that is the best predictor of a caregiver's relationship to her child. Main believes this finding has implications for the so-called constructivist approach to psychoanalysis. It has also been found that parents with highly unfavorable past experiences can in fact have securely attached children. Thus, Main views her research as leading to the study of how the psychoanalytic reworking of a life history toward greater coherence may hold a favorable outcome for succeeding generations as well as for individuals.
Early attachment research was criticized for countenancing a mind virtually unrecognizable by psychoanalysts. Often the mind and psychic structures were viewed as of no scientific importance. Attachment theory and research, with its emphasis on observable behaviors, was criticized also for overvaluing the significance of actual experience and underplaying the role of fantasy. More recent attachment research, however, has focused on intrapsychic “working models” consonant with psychoanalytic views concerning the acquisition of early self- and object representations. Attachment theorists such as Bretherington and Main are no longer working with reductionistic models that exclude mind and related processes. However, an artifact of hypothesis-driven research and scientific adherence to the principle of parsimony is that the psychoanalytic assumptions of overdetermination and multiple function have yet to find expression in the concerns of attachment theorists. Nonetheless, Main's research well exemplifies the new-look attachment research that is becoming increasingly relevant to the clinical and theoretical concerns of psychoanalysis.