The Adult Attachment Interview (AAI): Mary Main in a Strange Situation

Mary Main & Dan Siegel December-2010-UCLA[Updated 7-7-16]  Only 55% of us have “secure attachment”– a number which would worry us all if we knew about it–according to research on over 6,280  infant-parent pairs during 1970-1999. [FN1, 11]  And, the attachment we get as infants continues all our lives in our relationships, say related studies of over 10,500 adults done 1982-2009. [FN18]

The math says the other 45% of us suffer “insecure attachment.”  That means 45% can’t handle a committed, stable relationship with anyone, from childhood to the rest of our lives, as of 1999.  We also pass this emotional pain to our children, who turn out similarly.  A National Institute of Health article summarizes the secure rate:  “Infants with secure attachment greet and/or approach the caregiver and maintain contact but are able to return to play, which occurs in 55% of the general population.” [FN1, 11]

This is the blockbuster result of Dr. Mary Ainsworth’s 1970-1978 “Strange Situation” study of babies, as completed by her student Mary Main.  Main’s research led to shocking conclusions. (Dr. Main & Dr. Daniel Siegel, above)

Main discovered so many upset babies, that she got concerned about the parents. So in 1982 she created the Adult Attachment Interview (AAI) to study the adults, releasing results 1984-96.  Her journey was so “Strange” and involved it was published in language hard to decipher (or even google) for most folks. The tale took me weeks to unravel (footnotes below).

This huge “insecure” figure is a predictor of broken homes and broken hearts for half the nation. It starts to explain why we’ve got a 50% divorce rate. If  (like me) you’ve tried over-40 dating after divorce, it won’t surprise you that science shows 50% of adults out there can’t carry on a secure, committed, loving relationship. Ouch, you’ve experienced it.

And if 45% of us were “insecurely attached” in 1999, what’s the percent in 2016?  In 1999 most of us hadn’t heard of the Internet. In almost 20 years since, email, texting, and so on have further trashed our ability to relate in person. Several psychotherapists interviewed for this blog said that a round number of “about 50%” is a  conservative estimate for how many Americans lack secure attachment today. Many believe it’s much higher.

It gets worse; check out another “about 50%” shocker.  The 1998 Adverse Childhood Experiences (ACE) Study showed that two-thirds (64-67%) of  17,421 middle class subjects had one or more types of childhood trauma, and 38-42% had two or more types.  In less privileged populations, these numbers are far higher. A national average of all economic groups would likely show 50% or more suffer severe trauma from ACEs.

The ACE Study lists physical and sexual abuse and 8 other types, including traumas that happen even to newborns like physical and emotional neglect. [FN2]  Such trauma by definition puts children into technical “fight-flight,” a chronic state biologically proven to shut down the organism’s capacity for feelings of attachment and love. Think soldier in a  battle ramped up in “fight-flight”– he can’t really feel much love for the other side. [FN3]

And it doesn’t go away. Continued fight-flight puts the nervous system into freeze: our vagus nerve starts shutting down bodily functions, Dr. Stephen Porges shows. [FN4] So ACEs create the conditions underlying the top 10 medical causes of death in the U.S.

Half of us are in serious emotional health and medical trouble – and don’t even know it.  Let’s get informed –then we can heal. If we didn’t get securely attached as kids, we can develop “earned secure attachment.”  “It’s possible to change attachment patterns,” as Main said in 2010. [FN5]

 Strange Situation  Experiment

bowlby-johnAttachment Theory isn’t new, it just gets too little air time. British psychiatrist John Bowlby (left) developed it in the 1950s while working on the post-war orphan crisis. [FN6]  Bowlby believed that all infants would seek to stay close to parents, since such “attachment” promotes survival. In 1952 he published a study of toddlers’ responses to separation from parents. It showed that “when toddlers were placed in unfamiliar surroundings that provided no stable caregivers, they underwent three… stages of response to separation: protest, despair, and finally detachment.” [FN7]

Mary Ainsworth ca 1990Dr. Mary Ainsworth studied with Bowlby in London 1950-54, then researched his concept of “proximity-seeking behavior” in infant-mother pairs in Kampala, Uganda, published as “Infancy in Uganda” (1967).  Then she found “astonishing similarities” in Baltimore, MD pairs. [FN7]

Ainsworth created the Strange Situation in the early 1970s, as a science experiment at Johns Hopkins in Baltimore to document this infant behavior. “Ainsworth structured the Strange Situation to include three of Bowlby’s ‘natural clues to danger’… to arouse babies to seek proximity” to the parent, Main says. Researchers watch and video-tape through one-way glass, as infant-mother pairs react to apparent danger.  First a baby responds to a strange lab room; then to two entrances of a strange person; then to  two different separations from its mother. [FN8]

Babies were expected to stay close to parents as Bowlby thought.  Babies “that Ainsworth termed ‘secure,’ play and explore happily prior to separation; show signs of missing the parent during separation, such as crying and calling; seek proximity immediately upon the parent’s return; then return to play and exploration, ‘secure’ once again in the parent’s presence,” said Main. [FN8]

But 30% of babies did not act secure — they avoided mom.  They showed no preference between mom and the stranger!  “While a majority of infants behaved as expected and were termed secure, to Ainsworth’s amazement six showed little or no distress at being left alone in the unfamiliar environment, then avoided and ignored the mother on her return.”  [FN7]

Ainsworth decided to categorize these babies separately, as “avoidant” of mother. Now she had two types: (A) Insecure Avoidant, and (B) Secure. She concluded that their mom didn’t respond to them, or have the sensitivity to understand their actual need, so the infants felt “insecure.”  [FN8]

Still later Ainsworth saw that of the insecure babies, some had yet a third reaction: actually, they were “ambivalent” about mom. They were very distressed when mom left, but on her return, they alternated between avoiding and frantic clinging–plus, they never calmed down. Research showed that ambivalent attachment results from moms who are sometimes available, sometimes not, so babies learn they can’t depend on mom.

Thus it was that “surprisingly, Ainsworth found that infant responses to separation and reunion fell into three distinct, coherently organized patterns of attachment,” and added a third category: (C) Insecure Ambivalent, Main reports. [FN8-9]

By 1977 Ainsworth had categorized U.S. babies as (B) Securely attached 69%; (A) Avoidant 23%, (C) Ambivalent 8%.

By 1988 Strange Situation research had been done with 2,000 infant-parent pairs in 32 studies in 8 countries. By 1999 it  had been done globally with 6,282 infant-parent pairs. Some countries varied, but global results averaged the same as Ainsworth’s 1977 categories.  [FN1, 9, 11]

Scary Parents:  The Adult Attachment Interview

Mary Main BerkeleyIn 1973 Mary Main became Ainsworth’s grad student at Johns Hopkins in Baltimore, working on the Strange Situation experiments from their start. After her doctorate Main moved to Berkeley, to see if Ainsworth’s Kampala and Baltimore findings would replicate. [FN8]  In 1977 Main did a Strange Situation study of 189 Bay Area infant-parent pairs which did replicate Ainsworth. [FN8,10]

But by 1979 Dr. Main was making her own discoveries—out of concern about the parents. “In none of Ainsworth’s original observations was the possibility considered that some mothers… could also be frightening,” Main notes. “For my dissertation at Johns Hopkins, I watched 50 children in the Strange Situation… Using Ainsworth’s three-part classification (secure, avoidant and ambivalent), I found at least five infants could not be classified…

“I had visited most of the mothers in their homes,” Main said, “and I knew that at least three of the five mothers of the un-classifiable infants had behaved most peculiarly with their offspring. One—frighteningly, to me—had treated her toddler as an animal.” [FN8]

Ainsworth was concerned, too; in fact, she’d left some babies in her Secure set only because they didn’t fit her other two sets. [FN8]

By 1982, Main decided to “extend attachment theory to include the import of infant exposure to anomalous fear-arousing parental behaviors… The mother is the haven of safety that must be approached in times of danger. However, when the infant’s biological haven of safety has simultaneously become a source of fright, the infant is placed in an irresolvable and disorganizing approach-flight paradox,” she said.  [FN8]

Soon after, Main and Ainsworth agreed that 14-15% of babies actually belong in a new, fourth category: (D) Insecure Disorganized. Their mothers were so frightening that the babies couldn’t develop any consistent response at all. These infants “exhibited a diverse array of inexplicable or overtly conflicted behaviors in the parent’s presence” including “disorganization, disorientation, and confusion,” Main wrote. This included crying loudly then suddenly freezing; ignoring the parent to rock on hands and knees; moving away; raising hand to mouth in fear; or even swiping at the parent’s face. They seemed vulnerable to dissociation.   [FN11, 8]

This finding that a whopping 15% of average U.S. babies are so insecure they’re almost incoherent was so shocking, it was checked for years and not used as a formal 4th category until after 1990 (this delay confuses many scholars who look at 1982-1990). [FN12]  But the number held. Worse, among children of American adolescent mothers the rate is over 31%, and is over 25% in many Third World nations. [FN13]

Further: to remove the 14-15% of disorganized babies from Ainsworth’s original 69% Secure, reduced the Secure set to only 54-55%.  What was up with parents that 45-46% of their kids couldn’t manage secure attachment?  Main was developing  four categories: Avoidant 23%, Ambivalent 8%, Disorganized, 14-15%, Secure 54-55%.  [FN1, 11]

Mary Main '09 Bowlby-Ainsworth award(Inge Bretherton,Everett Waters)By 1982 Main had seen enough to create the Adult Attachment Interview (AAI).  She’d seen plenty of disorganized babies, and that meant a lot of scary parents. She needed a way to document the behavior of the parents. [FN14] (Right: Dr. Main (center) receives the 2009 Bowlby-Ainsworth Attachment Award.)

The AAI was designed to document the level of secure, loving attachment the parents had, during their own childhoods with their babies’ grandparents. Parents were asked “both to described their attachment-related childhood experiences – especially their early relations with parents – and to evaluate the influence of these experiences on their development and current functioning.”

Parents were studied not just on facts they gave, but on how coherent a narrative they could produce quickly. That’s easy for folks who had a secure childhood, but difficult for folks who did not. The AAI questions are designed to “surprise the unconscious” to yield information about the “state of mind with regard to attachment” that might not otherwise show up. Each AAI was taped and transcribed verbatim. Then transcripts were classified by specific patterns by independent trained specialists. [FN15, 10]

In 1982 Main also created the “Berkeley Longitudinal Study” to take 67 Bay Area infant-parent pairs from her 1977 Strange Situation and study them for a generation. In 1977 the babies were 12-18 months old.  Main and her team re-studied these pairs when the kids reached age 6 in 1982, and studied the kids again when they reached 19 in 1995.  [FN8,10]  In 1982 Main and her team gave three different tests to the Bay Area pairs:

— 1. The six year olds were again studied with their parents in the Strange Situation (as in 1977 when these kids were infants).

— 2. The “sixes” were also individually (without parents) given a new Separation Anxiety Test (SAT): they were shown pictures of children being separated from their parents, then asked how they felt. This was taped, transcribed and sorted by Main’s student Nancy Kaplan into Ainsworth’s three sets: secure, avoidant and ambivalent. [FN16, 8]

— 3. The parents were given the AAI (alone). Responses were sorted into three types of adult attachment matching Ainsworth’s three categories of infant attachment: Secure-autonomous (matching infant Secure), Dismissing (infant Avoidant)  and Preoccupied (infant Ambivalent). (Again, Main’s 4th category wasn’t used until after 1990.) [FN17, 8, 10]

Astonishing Results You’ve Never Heard

Mary Main, Erik Hesse '09 Bolwby-Ainsworth AwardDr Main’s first 1982 AAI results were so astonishing that attachment researchers have been buzzing ever since.  (Dr. Main and Dr. Erik Hesse, right.) Her results were also so important, it’s a tragedy that your doctor never learned this in medical school; your therapist (and mine) never heard of it; you’ve never heard of it; and so you have to read about it here, since the media doesn’t report it. It’s hard to even find this story on the internet; I had to sleuth it out.  Dr. Main’s initial AAI results were:

First: the six year olds’ 1982 responses to parents in the Strange Situation correlated strongly to their 1977 responses in the Strange Situation as infants — five years earlier. In 1977 the infants were Securely attached 69%;  Avoidant 23%, and Ambivalent 8%. In 1982 these kids at six responded to parents just as they had as infants, in the same percents. [FN7]

Second: the six year olds’ 1982 solo responses to the SAT photos also produced the same results and percentages.

Third: The parents’ 1982 solo responses to the Adult Attachment Interview correlated strongly with how their kids behaved as infants in the Strange Situation — five  years earlier.  The parents also turned out to be Secure (matching infant “Secure”) 69%;  Dismissing (matching infant “Avoidant”) 23%;  and Preoccupied (matching infant “Ambivalent”) 8%.

Fourth:  The match of the parents’ own 1982 AAI security, with how secure their babies were five years back in 1977, was at an unheard-of level. It correlated more than 70% of the time, in a field where a 20% correlation is highly significant. “A marked relation between a parent’s discussion of his/her own attachment history (AAI), and the offspring’s Strange Situation behavior 5 years previously, had been uncovered.”  [FN8, 10]

Fifth: When the 1977 infants reached age 19 in 1995, they too were given the Adult Attachment Interview. Again results correlated strongly: the 19 year olds’ responses in the 1995 AAI correlated precisely to their infant behavior in the 1977 Strange Situation, their 1982 behavior at six, and to their parents’ 1982 AAI responses.  This was astonishing.

It means that the Strange Situation predicts an infant’s emotional behavior for life, and the AAI proves it.  [FN8, 10]

Sixth: When after 1990 Mary Main’s four categories were used, the results of all these tests correlated just as strongly across the four new categories, as when data were sorted into only three sets. Main’s four categories did become Avoidant 23%, Ambivalent 8%, Disorganized, 14-15%, and Secure 54-55%.  Whenever responses to the Strange Situation, SAT, and AAI were sorted into these four sets, the same percents were found. [FN1, 11]

By 2009, over 10,500 subjects globally had been given the AAI and the results continue to average out the same.  [FN18]

“Researchers worldwide have replicated the relation originally uncovered in the Bay Area study between a parent’s status in the Adult Attachment Interview and an infant’s Strange Situation response to that same parent… The same average parent-to-child, secure/insecure match of 75% holds even when the interview is conducted before birth of the first child…

“Describing the strength of this relation across studies conducted several years ago, van I Jzendoorn (1995) calculated that it would take 1,087 further attempted replications, every one yielding insignificant results, to reduce the present relation between adult and infant attachment status to insignificance,” Main concluded. [FN7, p.1091]

Kathy’s news blogs expand on her book “DON’T TRY THIS AT HOME: The Silent Epidemic of Attachment Disorder—How I accidentally regressed myself back to infancy and healed it all.” Watch for the continuing series each Friday, as she explores her journey of recovery by learning the hard way about Attachment Disorder in adults, adult Attachment Theory, and the Adult Attachment Interview.


FN1  NIH: Benoit, Diane, MD, FRCPC, “Infant-parent attachment: Definition, types, antecedents, measurement and outcome,” Paediatr Child Health, Oct 2004; 9(8) p. 541–545 at:  4th subhead “Measurement” reports:
“The three ‘organized’ strategies (secure, avoidant and resistant) are assessed in the Strange Situation (SS), a 20-minute laboratory procedure where patterns of infant behaviour toward the caregiver following two brief separations are categorized… “Infants with secure attachment greet and/or approach the caregiver and may maintain contact but are able to return to play, which occurs in 55% of the general population… Infants with insecure-avoidant attachment fail to greet and/or approach, appear oblivious to their caregiver’s return… avoiding the caregiver, which occurs in 23% of the general population. Infants with insecure-resistant [ambivalent] attachment are extremely distressed by separations and cannot be soothed at reunions,  displaying much distress and angry resistance to interactions with the caregiver, which occurs in 8% of the general population.” [This NIH article earlier reports that the remaining “approximately 15% suffer insecure ‘disorganized’ attachment,” citing their own footnote which states “In normal, middle class families, about 15% of  infants develop disorganized attachment.” [23%+8%+15%  = 46% not securely attached.]

FN2  Felitti VJ, MD; Anda RF, MD, et. al, 1998, “Relationship of childhood abuse and household dysfunction to many of the leading causes of death in adults: The Adverse Childhood Experiences (ACE) Study,” American Journal of Preventive Medicine 1998;14:245–258.  Detailed article on the ACE Study:

FN3  Karr-Morse, Robin, Wiley, Meredith,  “Scared Sick,”  Penguin Basic Books, 2012

FN4  Porges, Stephen, PhD, “Body, Brain, Behavior: How Polyvagal Theory Expands Our Healing Paradigm,” 2013,
“Beyond the Brain: Vagal System Holds the Secret to Treating Trauma,” 2013,
–“The Polyvagal Theory for Treating Trauma,” 2011,
–“Polyvagal theory: phylogenetic substrates of a social nervous system,” International Journal of Psycho-physiology 42, 2001,  Dept. of Psychiatry, Univ. Illinois Chicago,

FN5   Earned secure attachment occurs when we experience harmful parenting, so we start with insecure attachment, but find ways to “rise above” childhood trauma and “are now securely attached… What’s more important than what happened to us, is how we’ve made sense of our own childhood,” Dr. Dan Siegel says. “When we make sense of our past… we become free to construct a new future for ourselves and for how we parent our children. Research is clear: If we make sense of our lives, we free ourselves from the prison of the past.”  (Source: video by Dr. Mary Main, Dr. Erik Hesse, Dr. Daniel J. Siegel, Dr. Marion Solomon: )
The question is: what ways?  How do we “rise above” and “make sense” of our childhood trauma?
“Mindfulness has been shown to be effective in healing insecure attachment,” say Siegel’s recent writings. “The purpose of both psychotherapy and mindfulness practice is to provide this internalized secure base. Attunement, whether it is internal in mindfulness, or interpersonal in attachment, is what leads to a sense of secure base.” (The Mindful Therapist: A Clinician’s Guide to Mindsight and Neural Integration, W.W. Norton, 2010;   Mindsight: The New Science of Personal Transformation,  Random House, 2010).
“The regular exercise of mindful awareness seems to promote the same benefits–bodily and affective self-regulation, attuned communication with others, insight, empathy, and the like–that research has found to be associated with childhood histories of secure attachment,” Siegel wrote earlier. “ Mindfulness and secure attachment alike are capable of generating… the same invaluable psychological resource: an internalized secure base.” (Siegel, 2007, Wallin, 2007, p. 5-6).
In his 1999 book “The Developing Mind: How Relationships and the Brain interact to shape who we are,” (Guilford Press),  Siegel defined “earned secure/autonomous attachment” as a pattern noticed by therapists doing the Adult Attachment Interview:  “individuals whose experiences of childhood… [were] likely to produce insecure attachment (avoidant, ambivalent, or disorganized),” but their AAI interview responses instead show “a fluidity in their narratives and a flexibility in their reflective capacity, such that their present state of mind with respect to attachment is rated as secure/autonomous. These individuals often appear… to have had a significant emotional relationship with a close friend, romantic partner, or therapist, which allowed them to develop out of  insecure…into a secure/autonomous AAI status.”

FN6  Bowlby, John, “The Nature of a Child’s Tie to His Mother,” British Psychoanalytical Society, London, 1958; “Attachment and Loss,” New York, Basic Books, 1969

FN7  Main, Mary,  2000, “The Adult Attachment Interview: Fear, attention, safety and discourse processes;” also titled “The Organized Categories of Infant, Child, and Adult Attachment: Flexible vs. Inflexible Attention Under Attachment-Related Stress,” Jour of Amer Psychoanalytic Assoc, 48:1055-1095; 2000.  *p.1091: “The same average parent-to-child, secure/insecure match of 75% holds even when the interview is conducted before birth of the first child…”

FN8  Main, Mary, 2005, with Hesse, Erik & Kaplan, Nancy, “Predictability of Attachment Behavior and Representational Processes at 1, 6, and 19 Years of Age – The Berkeley Longitudinal Study,” Chapter 10 of “Attachment from Infancy to Adulthood: The Major Longitudinal Studies,” edited by Klaus E. Grossmann, Karin Grossmann, and Everett Waters, pp. 245–304, New York: Guilford Press. Main refers to it as “Regensburg.” Regensburg 2005 .pdf
–Main’s summary of the Strange Situation in this document:
“Ainsworth structured the Strange Situation procedure to include three of Bowlby’s ‘natural clues to danger’ in eight episodes:  1. Introduction to the room.  2. Mother and infant are left alone in a toy-filled environment whose unfamiliarity supplies the first natural clue to danger.  However, the mother’s presence is expected to provide the infant with security sufficient for exploration and/or play.  3. Providing a second clue to danger, a stranger joins the mother and infant.  4. The mother leaves the infant with the stranger, providing two combined clues to increased danger. 5. The mother returns, and the stranger departs…. Many infants initially seek proximity but then, reassured of their mothers’ nearness, resume play. 6.  The mother leaves, and the infant remains entirely alone in the unfamiliar setting. Infant distress can be strong at this point, and this episode is often terminated rapidly. 7. The stranger, rather than the mother, enters the room. 8. The mother returns… By now, most infants are expected to be crying, and actively not only seeking proximity to mothers, but also… indicating a strong desire to be held… Nonetheless, they are expected to settle and renew interest in exploration and play by the end of this 3-minute period…
“Somewhat surprisingly, Ainsworth found that infant responses to separation and reunion in this procedure fell into three distinct, coherently organized patterns of attachment (“secure,” “insecure-avoidant,” and “insecure-ambivalent” (Ainsworth, Blehar, Waters, & Wall, 1978). … Given the length and complexity of this chapter, we suggest individuals divide reading to its three central parts (secure attachment, pp. 261–273; avoidant attachment, pp. 273–279; and disorganized attachment pp. 279–288).”

FN9  (“van Ijzendoorn 1988,” global replication of Ainsworth’s 3 categories; written before Main’s 4th category came into use) – van IJzendoorn, Marinus H.; Kroonenberg, Pieter M.  “Cross-Cultural Patterns of Attachment: A Meta-Analysis of the Strange Situation,” Child Development,Vol 59 No 1, Feb 1988, p.147–56.   Abstract:

FN10  Hesse, E., (2008) “The Adult Attachment Interview: Protocol, Method of Analysis, and Empirical Studies,” Chap. 25 of Cassidy, Jude &  Shaver, Phillip R. (Eds), “Handbook of Attachment: Theory, research, and clinical applications,” 2nd edition, 2008, p. 552-598, New York, Guilford Press. It was online (I downloaded it Aug. 2014) but was taken down or has web issues; try here:   [His 1999 version in Cassidy & Shaver’s 1st edition, Chap. 19,  “The Adult Attachment Interview: Historical and current perspectives,” p. 395-433 wasn’t online.]

FN11   (“van Ijzendoorn 1999” on Main’s new Disorganized category; confirms need for 4 categories)  van IJzendoorn MH, Schuengel C, Bakermans-Kranenburg MJ, Disorganized attachment in early childhood: Meta-analysis of precursors, concomitants and sequelae. Dev Psychopathol. 1999; 11:225–49.

FN12  Main, Mary, & Solomon, Judith, (1986), “Discovery of an insecure disoriented attachment pattern: procedures, findings and implications for the classification of behavior,” in Brazelton T, Youngman M. Affective Development in Infancy, Ablex, Norwood, NJ
Main, M., & Solomon, J. (1990). “Procedures for identifying infants as disorganized/disoriented during the Ainsworth strange situation,” in Greenberg, M. T., Cicchetti, D., & Cummings, M. (Eds.),. Attachment in the preschool years: Theory, research, and intervention (pp. 121-160), University of Chicago Press

FN13  “Among children of American adolescent mothers, the rate is over 31% (Broussard 1995). Disorganized attachment is also common among the Dogon of Mali (~25%, True et al 2001), infants living on the outskirts of Cape Town, South Africa (~26%, Tomlinson et al 2005) and undernourished children in Chile (Waters and Valenzuela 1999),” reports NIH [op cit FN1; NIH cites v. Ijzendoorn 1999 and the other authors just mentioned]

FN14  George, Carol, Kaplan, Nancy, & Main, Mary, “Adult Attachment Interview,” Unpublished MS, Department of Psychology, University of California at Berkeley, third ed. 1996.  Original 74-page MS dated 1984, 1985, 1996.  Described in FN10 op cit Hesse 2008. ( I have it, but it was removed from the web during 2014.)

FN15  Main, Mary B., “Adult Attachment Interview Protocol,” 11 pgs, 20 questions, no date or publisher. Dr. Main requires intensive training for use of the AAI. The questions per se, marked “Do not reproduce this material without permission of the author,”  are here:   UCLA’s Lifespan Learning Institute in Los Angeles holds AAI workshops and has an extensive CD  lectures on the AAI at

FN16  Kaplan, Nancy, (1987), Separation Anxiety Test (SAT): “Individual differences in six-year-olds’ thoughts about separation: Predicted to actual experiences of separation,”  Unpublished doctoral dissertation, University of California, Berkeley.
Kaplan, N. (2003, April), “The development of attachment in the Bay Area study: One year, six years, nineteen years of age.” Paper at the biennial meeting of the Society for Research in Child Development, Tampa, FL.

FN17  Main 1985; Main & Goldwyn 1984

FN18  Bakermans-Kranenburg MJ, van IJzendoorn MH, “The first 10,000 Adult Attachment Interviews,” Attach Hum Dev. 2009 May; 11(3): 223-63. doi: 10.1080/14616730902814762 at

More by Mary Main on the AAI:
Main, Mary, PhD, “Introduction to the special section on attachment and psychopathology: 2. Overview of the field of attachment,” Journal of Consulting and Clinical Psychology, 64, 237-243, 1996
Steele, Howard and Miriam, Editors, “Clinical Applications of the Adult Attachment Interview,” The Guilford Press,  New York, 2008

Comments are encouraged with the usual exceptions; rants, political speeches, off-color language, etc. are unlikely to post.  Starting 8-22-16, software will limit comments to 1030 characters (2 long paragraphs) a while, until we get new software to take longer comments again.

54,312 total views, 144 views today

Share Button

30 responses to “The Adult Attachment Interview (AAI): Mary Main in a Strange Situation

  1. Pingback: Adult Attachment Classification

  2. Pingback: Adult Attachment Interview

  3. Pingback: Adult Attachment Indicator

  4. Pingback: Adult Attachment Lab

  5. Pingback: Adult Attachment Disorder Nj

  6. After 20 years of trying to figure out why my husband can’t attach, this finally makes sense. Thank you for this great resource. Please send list of therapists in Orange County.

    • If we marry one, “we usually is one” (I was). Unless your hub wants therapy, the only way to get him there is for you to say “No” to bad treatment & go to therapy yourself. We can’t push others; we can only “model” mental health. I let my ex distance me for 27 years. I needed to say “No” and go to therapy for allowing bad treatment, before anything could change.

  7. Pingback: To Hide or to Cling |

  8. Thank you for this terrific info! This is very important. It links with (jargon-free) research I did and a resource website I have which you and others may find useful:
    Here’s a 2-minute read article about my research:

  9. Pingback: Articles on attachment theory | Joshua238's Blog


  11. Hello Kathy, Just a quick note to say how much I enjoy your writing and appreciate all your sleuthing on the subject of attachment and its history. – Marenka Cerny, MFT

  12. This excellent website really has all the information I wanted
    concerning this subject and didn’t know who to ask.

  13. Pingback: Mary Main in a Strange Situation: The Adult Att...

  14. Mary Ainsworth passed away. I believe Mary Main is at Berkeley. UCLA’s Lifespan Learning Institute in LA holds AAI workshops and has CDs on it:
    It would be nice to assess our own AAI score, but it’s dicey if people start assessing each other. It took me months to figure out how they discovered all this, as their writings are so obscure. Why? The more I study their documents, the more it seems they don’t want the average person using the AAI. Perhaps they are concerned lay persons might abuse the AAI to judge or classify each other? They insist that professional therapists get special training to use the AAI. Yes, all therapists should be getting training and using it! Yes, I’m upset most of them never heard of it. But us un-trained folks using it on each other? I wonder. Also check out Hesse, E., (2008) “The Adult Attachment Interview: Protocol, Method of Analysis, and Empirical Studies,” 2nd edition, 2008, p. 552-598 Guilford Press. try here: andMain, Mary B., “Adult Attachment Interview Protocol,” 11 pgs, gives 20 AAI questions, no date or publisher. The AAI questions per se, marked “Do not reproduce this material without permission of the author,” are here:

  15. One Love’s phone AP helps us (or our loved ones) assess our risk of being harmed in domestic violence:
    I wonder if we could decrease spanking and increase awareness of positive parenting with an AP to self assess your own AAI score. Parents seek support from family, friends, colleagues and the internet. They all could do an AAI to see if their loved one is at risk for problems attaching to their child. Try to download the One Love AP. It’s amazing! The Ravens gave $400K to support it. Whom can I contact? Can I reach Mary Ainsworth?

  16. Pingback: directory - Earn 50% per listing every month |

  17. I would like to alert your readers to my books – one in particular on infant development: Kenny D. T., “Bringing up Baby: The Psychoanalytic Infant Comes of Age,” (Karnac, 2012)
    It gives a detailed scholarly account of Attachment Theory:

  18. I am a practicing psychotherapist trained at one of only two CACREP accredited universities. I was trained on Bowlby, Ainsworth and the Strange Situation; so yes if you happened to be my client, your therapist was indeed trained.

  19. Joseph Maizlish, MFT.

    Among the 45% lacking secure attachment are many who have learned to cope & carry on a good relationship by, for example, mutual recognition of special needs. Thank you for tracking the history of Ainsworth’s categorization and Main’s further development. I hope you get to present this in written and other forms to many lay and professionals.
    I hear echoing in my mind Dan Siegel’s remark that “The most robust indicator of balanced functioning of the young child’s brain is the coherence of the self-narrative of the principal caretaker.” Siegel’s remark was about the “SELF-narrative,” not the attachment narrative – but those must go together. Good relationships within a coherent-self caretaker and within a child and between them makes likely good relationships with others much more likely. From that we also get what Main called the “coherent narrative” regarding the parent’s attachment to their parents, as you note.
    What a high correlation between the parent’s attachment to their own parents, and their kids’ attachment to the parents! So these factors are shown to be moving through a lifetime and across generations.

    • Ever since I discovered this material, I’ve been doing my utmost to get that “earned secure attachment.” It’s so difficult re-wiring our brains; hope I’ve made a little progress in 5 years…Thank you for noting that this was a tough story to piece together. I get queries every day from all over the country saying therapists like you and Sue are hard to find.

  20. Thank so much, I’m eager to have your article! I’ve done 5 or 6 blogs on Dan Siegel and one on Allan Schore at:
    Have a look through my blogs at
    I was at Dan’s UCLA conference in March 2013 where the new campaign Relationships First was created (I signed the email list but never heard from them.) I’d love it if you could put an update on Relationships First in your blog for my site, so we could promote it and give readers an update!

  21. Hi Kathy, I would love to help! Dr. Dan Siegel and Dr. Allan Schore are rock stars in this arena. I want to recommend Dr. Stan Tatkin, who studied with Dr. Schore, and his Psychobiological Approach to Couples Therapy (PACT) informed by attachment studies and neuroscience. Fantastic stuff, we have to get couples/parents feeling secure so they can provide a secure attachment experience to their children. They have a new campaign called Relationships First which Dr. Siegel and Dr. Tatkin are founders (along with Harville Hendrix and other heavyweights) with the intention of highlighting the importance of relationship (attachment) for healthy living. Yes, I would love to submit an article with my insights regarding finding a good attachment therapist. I will be in touch, and thanks again for your site.

  22. “However, her results were also so important it’s outrageous that your family doctor never learned about this in medical school; your therapist (and mine) never heard of this…”
    Love your site, love it! I understand your frustration with the lack of knowledge some in the therapeutic community display about attachment and trauma. But, NOT ALL OF US! Ainsworth’s Strange Situation is taught in masters level counseling psychology courses. Any psychotherapist should at least know about it. I’m a psychotherapist interested in attachment and a trauma informed approach and I work with many others in the field who are likewise. I fear readers may eschew therapy and the empathetic, healing connection it offers because of some bad apples. We know it’s the empathic connection between the therapist and the client that informs healing, not the cognitive based exercise they take home. Therapy can and has been successful for many no matter what the approach, as long as there is a therapeutic alliance and empathic connection. There are some really good therapists! Talented, heart centered, empathic, well educated therapists and social workers and doctors who connect at a heart level. True, our culture has been organized around a model that emphasizes left brain answers to right brain problems for a long time. The good news is we are all going through a paradigm shift together. Thank you again for your site, it is informative and much needed.

    • I am delighted to hear that Ainsworth’s Strange Situation is taught in masters level psychology courses and practicing psychotherapists should know of it.I had to go through three therapists who were ignorant of these principles, and damaged me to the point that I quit therapy, before I did the research myself and learned all this. See my book chapters starting with “I Oughtta Have My Head…
      I’ve met too many people who have had the same experience.
      BUT OF COURSEe no one can heal without good attachment therapists such as yourself! Thus my book title! It’s far to dangerous to “do it yourself!”
      Where you could really help would be to write a piece for my website on how to find a really good attachment therapist! I get queries on this every day from all over the country.

  23. Pingback: Developmental Trauma: What you Can’t See CAN Hurt You | "Don't Try This at Home"

  24. Pingback: The Hole in Me | "Don't Try This at Home"

  25. Pingback: The Hole in Half of Us | "Don't Try This at Home"

Leave a Reply

Your email address will not be published. Required fields are marked *

 characters available