Attachment Theory is controversial but it may be here to stay.
Before attachment theory, psychotherapy was in a state of fragmentation and conflict. Different schools of thought had identified pieces of the puzzle, but no group had an overarching theory that provided satisfactory answers for how people become the way they are. The field was a battleground of ideas and warring, mutually disdainful professional groups. Psychoanalysts, perhaps the most superior group, who worked with free association, disapproved of family therapists’ attempts at reorganizing the chaotic lives of parents and children. The emotionally focused psychodynamic therapists derided the cognitive behavioral therapists as too simplistic.
Self-reported study results did little to settle the score. According to a survey published by Consumer Reports in 1995, a large percentage of psychotherapy patients improved, no matter what their therapists’ theoretical orientation. The quality of the therapeutic relationship seemed to be the critical factor, rather than any particular school of thought.
Today the findings of psychological research, buttressed by neuroscience studies, have validated attachment theory, which posits that successful relationships of all kinds—with children, parents, friends, and partners—depend on how we, as infants and toddlers, bonded with our original caretakers. It has also encouraged therapists, many of whom had been trained to maintain a level of professional detachment, to form stronger personal bonds with their patients. More importantly, it has given patients themselves—even many severely traumatized ones—reason to believe that positive change is possible through therapy.
Attachment theory has its detractors, who say the theory’s strength is also its weakness: It is too all-encompassing, more useful as a map than a guide. And clinicians like me are uncomfortable with the way it can perpetuate blaming mothers. But there are good reasons for its considerable influence, inside and outside psychotherapy. What John Bowlby, the originator of attachment theory, called our initial “affectional bonds” provide a template for our worldview, our interpersonal relationships, even our brain structure and chemistry. In a profound and poetic way, these bonds shape the narrative of our lives.
Although attachment theory itself was first developed in England, it was in the United States that research protocols were established to assess the attachment style of infants. At Berkeley, groundbreaking work was done on assessing adult attachment styles. But it was pioneering neuroscience research done at UCLA that turned a theory into a powerful mainstream clinical movement.
Two-year-old Laura, needing a minor operation, spends 8 days in the hospital. Her parents are allowed to visit her every other day. On their first visit she bursts into tears and begs them to take her home. Eventually she appears to quietly accept the situation, not even waving goodbye to her mother after their final visit. When her mother returns to take her home, Laura is withdrawn and mistrustful. Although she seems more cheerful after a couple of days, she goes through a period of sleeping badly, is irritable and distressed, refuses to let her mother out of her sight, and has temper tantrums.
—(Robertson, 1953a, 1953b; Robertson and Bowlby, 1952)
The importance of the mother-child bond may seem obvious now, but it was a revolutionary idea when first proposed. John Bowlby (1907–90), a mid-20th-century British psychoanalyst, integrated concepts from Darwin’s theory of evolution and Freud’s developmental framework to formulate the concept of “attachment.” Slowly, over 30 years, Bowlby began to identify attachment patterns. At the London Child Guidance Clinic he first noticed that parents’ unresolved childhood traumas precipitated emotional disturbances in their own children. Bowlby’s work with World War II orphans and child evacuees led him to conclude that the lack or disruption of primary attachment between parent and child has lifelong effects, including depression and a minimal sense of self.
At the time, mother-child bonding went against mainstream Freudian psychoanalytic theory, which focused instead on exploring the Oedipus complex (the 3-year-old child’s interest in the opposite-sex parent). Otherwise, the general view in the 1930s was that the infant attached to the mother because she was the source of food and comfort. By the 1950s, mainstream psychoanalysts had concluded that the child’s early relationship to the mother was crucial. Erik Erickson, in particular, theorized that the infant’s relationship with the mother regulated the basic functions of feeding, sleeping, and eliminating, as well as creating “basic trust,” the foundation for the child’s sense of identity.
Bowlby’s approach provided an overarching theory for why this should be so. Humans, like all other mammals, Bowlby concluded, have a powerful, innate, biologically based drive to stay close to the parent for physical, emotional, and psychological safety, thus maximizing the possibility of survival for the infant and the species.
“To remain within easy access of a familiar individual known to be ready and willing to come to our aid in an emergency is clearly a good insurance policy—whatever our age,” he later wrote in A Secure Base: Clinical Applications of Attachment Theory, one of the trilogy of books (Attachment, Separation, and Loss) he published between 1968 and 1988.
Unlike Freud, who only welcomed new ideas he could assimilate into his own theories, Bowlby not only integrated ideas from other academic disciplines but was interested in research and replicable results.
A mother and toddler enter a room filled with toys. The child explores the toys, shows them to her mother and then settles down to play. Three minutes later, a young woman joins them. The mother soon leaves the room and the toddler protests vociferously, clinging to the doorknob, crying for the entire two minutes she is away, and refusing the stranger’s help. When the mother returns, the child immediately welcomes her. The mother picks her up, comforts her with a hug and talks to her in a calm reassuring voice. The child returns to playing on the floor. A few minutes later, both the mother and the caregiver leave the room, and the child, now alone, is even more bereft than before. When the mother returns a few minutes later she once again easily reassures and comforts the child.
—Patterns of Attachment: A Psychological Study of the Strange Situation, 1978 (M.D.S. Ainsworth, M. C. Blehar, E. Waters, S. Wall)
In the 1970s, Margaret Ainsworth, Bowlby’s longtime collaborator, developed the Infant Strange Situation designed to assess 12- to 18-month-old children’s attachment styles with their main caregivers, usually mothers.
During the 30-minute protocol, children and caregivers are put into a series of stressful separations and reunions like the ones described above. How the child responds to the mother’s return (technically known as reunion behavior) is evaluated to assess the child’s connection to the parent. Out of those studies, Ainsworth identified three attachment styles: secure, avoidant, and ambivalent.
Before her research, it was generally believed that children who protested were “difficult” babies and that quiet children were “good” babies. But Ainsworth’s study demonstrated that a secure child is the one who behaves as the little girl did above—she is happy to be in the presence of her mother, who encourages her to explore the toys, but is quite upset when left alone or with the strange young woman, and easily soothed by the mother when she returns.
Good quiet babies, judging by cortisol and adrenaline levels found in saliva samples analyzed later, are the ones in trouble. The avoidant baby withdraws from its mother and denies or minimizes distress. He does not approach the stranger. When the mother returns, he ignores her and doesn’t express relief or even interest in her.
The ambivalent (sometimes termed “anxious”) child seems stuck between staying close to the mother and exploring the toys. She is quite distressed when the mother leaves and doesn’t like being alone or with a stranger, but does not run to the mother when she returns.
When doing her Strange Situation research at the University of California, Berkeley psychologist Mary Main, who had been a graduate student of Ainsworth, always found a number of children who were not easily classifiable. In 1990, Main and graduate student Judith Solomon, Ph.D. ’86, identified a fourth attachment style—the disorganized (or disoriented) category. Disorganized attachment arises from the child’s irresolvable predicament in which the parent is both frightening and unable to offer the child any protection or safety.
In their Strange Situation studies, Main and Solomon found that a disorganized child, with a dissociated inattentive disorganized mother, was the only one of any of the groups who would actively engage the stranger when the caregiver left the room, only to scream aggressively and then retreat and cry when the mother returned.
Robert Karen, whose comprehensive 1994 book Becoming Attached introduced attachment to both clinicians and the general reader, commented that “once researchers had these … attachment categories, all sorts of questions previously confined to theoretical speculation were suddenly accessible to empirical study…. In the coming years, psychologists would use the Strange Situation to correlate the child’s attachment style with character development, schoolwork, problem solving abilities, self-reliance, self-esteem, peer relations, (and) general sociability.”
Choose five adjectives that reflect your childhood relationship with your mother. I (the researcher) will write them down, and when we have all five, I’ll ask you to tell me what memories or experiences led you to choose each one. (Repeat for father)
To which parent did you feel closest and why?
When you were upset as a child, what did you do, and what would happen?
—Adult Attachment Interview, 1984 (Carol George, Nancy Kaplan, Mary Main)
Curious whether the parent’s attachment status mirrored their child’s, Main and her graduate students and coauthors (Carol George, M.A. ’78, Ph.D. ’84, and Nancy Kaplan, M.A. ’84, Ph.D. ’87) evolved an interview method, the Adult Attachment Interview (AAI). This was a way of assessing adult attachment, a major development in moving the field beyond the study of infants.
The semistructured hour-long interview measures adults’ “states of mind about attachment.” The AAI asks participants to articulate and reflect on attachment-related memories while simultaneously maintaining a collaborative, coherent conversation with the interviewer. It contains 20 survey questions, 3 of which are reproduced just above. According to Erik Hesse ’81, Main’s collaborator and husband, participants are asked about experiences with parents and other significant caregivers, notable losses and trauma, and, if relevant, experiences with the subject’s own children.
The complex scoring process assesses the coherence or incoherence of the respondent’s narrative. Whether early experiences are favorable or difficult, respondents are seen as secure if they can give a coherent, believable account that accurately describes the feeling of their early attachment experiences. Avoidant adults provide a more limited account of their early bonding experiences, often can’t recall memories, and minimize or rationalize negative experiences and painful feelings. Anxious or ambivalent individuals give long-winded and discursive answers that are hard to follow. They may digress into idealized, negative, or ambivalent descriptions of early experiences. They often “don’t know,” and exaggerate (for example, “terribly sad,” or “very, very heartbroken”).
These styles correspond to Ainsworth’s three infant categories. Main also included a category for people who become disorganized and dissociated during memories of loss, abuse, and trauma, finding it difficult to remember, think, or collaborate with the interviewer.
The AAI is probably Main’s most important contribution. Marion Solomon, director of UCLA’s Lifespan Learning Institute, who organized a UCLA attachment conference in 2008 devoted to Main and the AAI, said that because Main’s original interest was in language, “she was able to find out not from the words or the stories,” but from how people told their stories, “what must be going on at a much deeper level, a brilliant finding.”
The AAI was tremendously influential, sparking a wave of other adult attachment studies. Of all available measures, including personality assessments, socioeconomic background, and intelligence, the AAI is the strongest predictor of infant attachment to parents with a 65 to 80 percent correlation between the child’s Strange Situation classification and the parent’s AAI category, according to Dan Siegel in his book The Developing Mind. Administered to mothers in the last three months of pregnancy, the AAI frequently predicts accurately a child’s attachment style at 18 months and again at age 18.
In a 2009 study published in Neuropsychopharmacology, first-time mothers, whose attachment styles had been identified during pregnancy with the AAI, were put into an fMRI machine when their babies were 11 months old. They were shown multiple photos of their babies and unfamiliar babies expressing a range of emotions from happiness to distress. The reward centers in the brains of securely attached moms lit up when they saw their own baby’s faces—even when their baby’s face was sad. With avoidant mothers, the centers of the brain associated with unfairness, pain, and disgust are activated by the unhappy baby. Blood tests support these findings.
—”Adult Attachment Predicts Maternal Brain and Oxytocin Response to Infant Cues”
Neuropsychopharmacology, 2009
(Lane Strathearn, Peter Fonagy, Janet Amico, P. Read Montague)
Attachment theory might have remained a series of elegantly explicated hypotheses had it not been for the pioneering work of two prominent neuroscientists at UCLA—Allan Schore, Ph.D., and Dan Siegel, M.D. Where an earlier generation of researchers (Ainsworth and Main) had concentrated on assessment, Schore and Siegel focused on integrating attachment theory with the neurobiology of brain development and clinical practice.
They worked across academic disciplines as diverse as medicine, sociology, neuroscience, and developmental psychology. They drew upon multiple studies, such as the one cited just above, that used the new technical advances, such as the MRI, to confirm Bowlby’s theories. This created a new paradigm, which Siegel calls interpersonal neurobiology.
Siegel is probably the person who has made attachment theory and neuroscience the most accessible to the ordinary person, because he has the rare capacity to translate scientific concepts into everyday terms. He has written a number of books for the technical reader (The Developing Mind), for parents (Parenting from the Inside Out), as well for the general reader (The Pocket Guide to Interpersonal Biology). Still, he readily acknowledges the debt his work owes to Professor Main.
“Mary Main is a very, very important figure in child development and parental research,” said Siegel. “She has illuminated in a beautiful, beautiful way the importance of narrative.” The AAI findings, he said, provided “a pivotal moment” in his own development, as did Main’s concept of “earned secure attachment.” And they inspired him to study both attachment theory and neuroscience.
Although Siegel is still conducting research, supervising Ph.D. students, and seeing patients, he seems to be writing and lecturing primarily about Mindsight, his term for self-awareness. In his books and TED Talks, he explains what he means by interpersonal neurobiology. It conceives of the brain as a “social organ” that is constantly reorganized by social relationships. Earned secure attachment fits nicely into his worldview. According to Main, adults can change their status from insecure to secure through effective therapy or solid intimate relationships. Securely attached adults are self-aware, can empathize, and feel connected to others—which, says Siegel, not only develops resilience, but also shapes the brain’s physiological structure by expanding neural networks (for example, between the right brain and the left brain). Indeed, he told me, “you can make a case that a coherent AAI narrative reflects an integrated nervous system.”
Allan Schore, who since 1996 has been an associate clinical professor at UCLA’s Center for Culture, Brain, and Development describes himself as a “clinician-scientist.” Schore has recast attachment theory into “a relational and regulation theory.” He also teaches and lectures all over the world. Spending a Saturday afternoon in his Berkeley study group is like standing under a bracing intellectual waterfall.
Schore has written three immensely detailed, technical books, fascinating but hard going even for the technical reader: Affect Regulation and the Origin of the Self, Affect Dysregulation and Disorders of the Self, and Affect Regulation and the Repair of the Self. He has published a fourth just this year: The Science of the Art of Psychotherapy.
For Schore, too, “The essential task of the first year of life is the creation of a secure attachment bond between the infant and the primary caregiver.” The good news for the parents of the world, according to Schore, is that even when they get it wrong, they can regroup and help the young child recover from distress and manage feelings.
The attuned parent can actively help the wet, tired, cold, hungry infant feel secure by responding with kind eyes, a relaxed body, gentle hands, and a soft voice. Over time, the child begins to manage difficult emotions and situations and to develop resilience—which, as Schore points out, “is an ultimate indicator of attachment security.” This is what Bowlby understood more than a half-century ago.
Schore, like Bowlby, views attachment as an evolutionary process in all mammals, the function of which, especially in humans, is to drive brain development. How mothers touch, comfort, look at, respond to, and talk to their babies all affect how the baby’s brain develops. Schore believes, as does Siegel, that as the mother helps her baby manage emotions over time, the child’s neural networks, the actual scaffolding of the brain, are established. This supports the child’s later capacity for emotional intimacy, empathy, self-awareness, and the foundation of a coherent sense of self. Or not.
When a baby’s ongoing relationship with the mother is disturbed by loss and separation, her brain does not develop optimally and she is left with the experience of uncontrollable feelings and behavior. As in the case of Maggie.
Maggie was 6 months old when her mother went into the hospital for many weeks. Maggie is now an angry and defiant 10-year-old having difficulties in school. Her parents are worried about [her] approaching adolescence, and as a result, her father has become more authoritarian and demanding, setting up more rigid rules and consequences. These have backfired, exacerbating Maggie’s disorganized attachment and creating significant conflict between the parents.
In the interests of full disclosure, I have to confess that I am not an objective or disinterested observer. I have been a psychotherapist in California for more than 30 years, in a period of enormous expansion in psychotherapeutic theory and practice. Initially, I was trained in psychodynamic therapy (a kissing cousin of psychoanalysis) and family therapy, but then embraced other modalities—Gestalt, psychodrama, group therapy, and somatic approaches. I am not a pure attachment theorist or therapist; as far as possible, I utilize what works for my clients, integrating everything of importance learned over the years. Plus, I pay attention to temperament; the family system; birth order; ethnic, cultural, and religious influences; trauma; and anything important that shows up.
I also like to bear in mind the hopeful and grounded empathic belief of D.W. Winnicott, another British mid-century psychoanalyst: There are no perfect mothers, but there are “good-enough mothers.”
What is most interesting to me clinically is Mary Main’s concept of “earned secure attachment,” which I believe extends to the therapist-client relationship. If a strong, positive bond is established, the “good-enough” therapist will help the client identify and manage emotions, yet not be overwhelmed by them; develop empathy for himself and others; expand self-awareness; and learn to calm down and reach out to others.
In my own work with families, I focus on helping parents reorganize their attachment relationships with their children. For example, I recently worked with Maggie and her family (whose identities have been obscured). When I first saw Maggie, she was alternating between bouts of hyperactive play with her well-adjusted younger brother, and complete withdrawal into computer games, a classic example of Bowlby’s description of early maternal deprivation. And the parents could not agree on a strategy, which was putting strain on the marriage.
I coached the father on softening his expectations, as well as his eyes and voice, teaching him the empathy and kindness that he had not received from his father. Maggie responded well. She spent a good amount of time in our sessions first leaning up against her parents, often ending up with her head in her mother’s lap, gazing up at her, while both parents talked to her and stroked her face.
Then, like the Strange Situation babies, she would head out to explore my toy collection with her brother, often returning with a toy or inviting the parents to play with her on the floor. I encouraged the parents to follow her lead with this at home, and in spite of the father’s continuing worry about “spoiling” her, she began to calm down and seek their help when she needed them. The family went on vacation during the summer; when they returned, the mother called to say that their trip had gone very well. They had all enjoyed themselves, and Maggie was much “calmer and easier.”
So where does this leave us? On some level, the theories of Bowlby, Ainsworth, and Main, now validated by neurobiology, simply tell us what we have long intuited—that the quality of early relationships organizes our experience of others and ourselves. We now have a coherent narrative about the process of becoming who we are; a more complex understanding of all our relationships, with fMRI studies to prove it; and a new, yet familiar twist on psychotherapy. Nature and nurture now seem profoundly interconnected.
Frances Verrinder is a licensed marriage and family therapist in San Francisco, where she practices individual, couples, and family therapy. She also teaches romantic attachment workshops at Esalen Institute. She grew up in England but married into a family of Cal alums, including mother–in-law Evie Griffith ’32, sister-in-law Jean Stice ’62, husband Michael Griffith ’65, and stepson Lucas Griffith ’07.
