Attachment theory through a cultural lens

Father and child (photo by Fredrik Lidarp), Australia

Father and child (photo by Fredrik Lidarp), Australia


In an article titled “Attachment and Culture (citation below),” Heidi Keller exposes attachment theory’s Western, middle-class assumptions. She argues:

… the definition of attachment in mainstream attachment research are in line with the conception of psychological autonomy, adaptive for Western middle-class, but deviate from the cultural values of many non-Western and mainly rural ecosocial environments.

Keller shows how attachment theory, particularly research that follows on the heels of John Bowlby’s original theory and Mary Ainsworth’s Strange Situation Procedure, assumes the most formative attachment relationship occurs between a mother and her infant. (For a further discussion of Bowlby’s and Ainsworth’s work, see my post, Let There Be Love!) But Keller points out that the primacy of the mother-infant bond for attachment may only be the norm “in Western middle-class families which compose less than 5% of the world’s population.”

In most cultures and socioeconomic groups, limited

resources and daily survival needs require distributing caregiving responsibilities across a network of relatives, including aunts, uncles, grandparents, and siblings (technically called alloparenting). And although Keller doesn’t include in her study daycare centers and other forms of hired childcare workers, they also play significant roles in rearing infants for many working parents.

Even when attachment is narrowly defined and studied as occurring primarily between mother and infant, the distribution of Ainsworth’s core attachment styles differ widely across cultures. According to Ainsworth, there are three dominant attachment styles:

  • Securely attached: The infant has a sensitively attuned mother, which supports the infant in feeling safe exploring the environment as well as seeking solace and safety from mother.
  • Insecurely avoidant: The infant has experienced a loss of attachment, or mother is misattuned to the infant’s needs and expressions, which leads the infant to respond as if efforts at seeking safety or solace will go unheeded. One common consequence is inhibited emotional and physical expression.
  • Insecurely ambivalent: The mother is often inconsistent in her parenting style. These infants showed the greatest distress when mother would leave, and were also unable to explore freely when mom was present. Infants may alternatively cling to mom and push her away after she returns (resistant), or react passively as if overwhelmed by distress (passive).

Although Ainsworth’s results showed a relatively stable distribution of these three attachment styles, Keller shares research that reveals a different picture of both the proportion of each style within populations as well as alternative explanations for the different styles:

In Northern Germany [some researchers] … replicated the Ainsworth Strange Situation with 46 mother-infant pairs and found a different distributions of attachment classifications with a higher number of avoidant infants: 52% avoidant, 34% secure, and 13% resistant. … The Japanese case is another example. [Researchers] … studied 60 pairs of Japanese mother-infant pairs and compared the Japanese distribution with Ainsworth’s distributional pattern. There were no significant differences in proportions of securely attached (68%) and insecurely attached (32%) infants [like Ainsworth’s results]. However, the Japanese insecure group consisted of only resistant children, with no avoidant ones… . Finally, there is the Israeli case with the … study that also revealed a high frequency of the ambivalent pattern. [The Northern Germany researchers] … interpreted their findings as expressing a greater parental push toward children’s independence, whereas the Israeli kibbutzim and the Japanese data were interpreted in terms of underexposure to strangers.

Bowlby’s theory of attachment was originally lauded for its evolutionary approach to infant development and its focus on the significance of the attachment bond for creating a safe base in potentially dangerous environments. His work has also been foundational for understanding the nature of grief and loss, and has been expanded to look at neurobiological aspects of these experiences, including their impact on heart rate and cortisol release. Furthermore, Ainsworth’s attachment styles have been correlated with later adult attachment patterns.

Nevertheless, as Keller stresses, from a Neo-Darwinian perspective, context is a determining factor for what constitutes an adaptive attachment style, but this crucial component has often been ignored by researchers of attachment theory. Keller wrote:

“Neo-Darwinian evolutionary theory … claims that contextual information is crucial for defining adaptation, thus putting variability in the very centre of evolutionary theorizing. The core assumption is that individuals need to select the behavioral alternatives that promise the highest reproductive outcomes in a particular ecological situation.

With this interpretation of evolutionary adaptation, Keller explains:

“Therefore, secure attachment is not ‘better’ than insecure attachment but a different way to maximize reproductive success.”

This is essentially the attitude taken by trauma-informed approaches to psychotherapy such as sensorimotor psychotherapy, which supports clients in both honoring how they survived maladaptive circumstances while also developing more secure patterns of attachment/relationships.

Furthermore, as anyone who has had to overcome early childhood abuse or neglect can attest, the impact of early attachment extends beyond adaptation to include one’s value system and core sense of self. And when the culture maintains models for ‘ideal’ attachment, such as preferencing the mother-child bond, these models can inadvertently be pathologizing if your childhood was less than ideal. As Keller noted,

Indeed, security of attachment is not simply a behavioral category; it is also a moral ideal in as much as it provides a pathway to the development of culturally valued qualities, such as self-confidence, curiosity, and psychological independence.”

I would also add that the fantasies and images we hold about attachment, such as preference for the maternal-child bond, can contribute to undervaluing other avenues to meeting attachment needs that are just as beneficial to infant development. These other relationships are obviously important to developing infants, but they are also significant for healing from histories of childhood abuse and neglect for which identifying positive relationships in one’s past can become internal models for building future capacities and relationships.

[Citation: Keller, Heidi. “Attachment and Culture.” Journal of Cross-Cultural Psychology, 2012, XX(X) 1-20.]

© 2013 Laura K Kerr, PhD. All rights reserved.

Laura K. Kerr, PhD, IMFT is a mental health scholar and registered marriage & family therapist intern in the San Francisco Bay Area. For more information, visit her website.

11 responses

  1. While I believe that cross-cultural studies are very needed to make generalized theories on human behavior, I do not think that the variety of parenting approaches by culture and socioeconomics dismisses the decades of research that supports Attachment Theory. If that were the case, we could toss away the mounting evidence against the psychological effects of corporal punishment that take place in other parts of the world outside the West.

    What’s also important to know is that Ainsworth’s research developed from watching caretaking habits from tribal women in West Africa, not suburban white women. When finished with her research in Africa, she moved to the states. No surprise, the results on child behavior was the same regardless of where the child was born.

    With that also being said, it’s unsupported to say that “insecure attachments are not inferior to securely attachment styles.” We have lots of evidence supporting the relationship between attachment style, and health outcomes for children-future-adults. Such studies show that people with insecure attachment style have higher levels of toxic stress that makes them prone to anxiety, depression, PTSD and riskier behavior. And yes, like childhood adversity (like divorce, separation, parental mental illness, ect.) it has physiological health effects as well…

    Regardless, of socioeconomic…

    and Race…

    Really, it can seriously be argued that having an insecure attachment outcome with a primary caregiver (and secondary caregivers such as grandparents, family members, father), could be counted as an ACE, considering the physiological and health repercussions.

    Lastly, I find it very confusing that you dismiss or challenge the stress of the maternal-child bond in Attachment Theory. Cross-culturally speaking, in virtually all societies outside the west, mothers are the primary caregivers. Even more so, there is science showing as early as minutes postpartum, infants stress when away from their mothers, and overwhelming prefer their mothers’ touch, proximity, smell, and voice to sooth them, (even other their fathers).


  2. Boy, I would like to see much more animal and brain research on this before heading across human environments. We are just talking about basic mammal medical physiology. How many important cultural differences are there in other organs of the body – heart, kidney, etc.? Not many. Why should brain be different? Why should the fundamentals of brain-behavior infant-parent be that different for any primate/mammal/animal? Likely, not much as well.


      • Well, it’s simple logic and basic medical biology. Bowlby’s original attachment work was done on a specific monkey but know it’s known the species he used has maternal-infant bonding behavior different from others. Lots of work still to do before this model/theory will be fully worked thru – or not.


      • The field of attachment is a large one, and quite a bit of research has been conducted since Bowlby’s original work. Furthermore, Ainsworth is another key contributor to the field, and again, much more research has been conducted in response to her original work, which Keller’s article reviewed. A good resource for seeing further developments, including in the field of neuroscience, is David Wallin’s Attachment in Psychotherapy.


      • Yes, but we do not yet have the full brain/hormonal/neuronal modeling and theory testing needed in brain science and animal ethology. Ainsworth literally knew nothing about behavior and the brain his studies and ideas were solely observational, as are most clinical claims, not experimentally proven – at all.

        Claims made on observational information is just anecdotal “evidence.”


      • Neufeld ( Canadian psychologist ) also has much neuroscience to back his paradigm. He and Dr. Gabor Mate ( Canadian physician ) both come at it from a developmental attachment foundation so conceptually and physiologically they unite. Together they make tremendous sense. Unfortunately the ground on which they walk counters much of today’s cultural beliefs about what a child needs and how we can best provide for them to reach their ultimate developmental goal of maturation.


      • You’ve definitely got me interested.

        I think it’s helpful to at least question normative practices for child-rearing, especially given the prevalence of ACEs, and that their impact on development still needs to be explored. The concern for what counts as “normal” probably ought to be replaced with the question, “What counts as humane and adaptive?”

        Thanks again for sharing.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: