Childhood Friends, Not Moms, Shape Attachment Styles Most

2025-11-1415:11283103nautil.us

Childhood Friends, Not Moms, Shape Attachment Styles Most: A new study upends conventional wisdom about how we relate to those closest to us.

Humans are social animals. We depend on our friends, partners, and family members to steer through troubled waters and cheer us on when we shine. One popular school of psychology known as attachment theory suggests that these close relationships tend to follow established patterns that differ from one person to the next: Some of us feel secure in our relationships, while others are more anxious about abandonment, less willing to trust even those we hold most dear.

Nautilus Members enjoy an ad-free experience. Log in or Join now .

Now a large, new, 30-year study has found that our earliest friendships may have the biggest impact on how well we “attach” to friends and romantic partners in adulthood. If true this finding would upend conventional wisdom that our relationships with our parents leave the biggest mark on our attachment styles later in life. The team of researchers found that, in fact, mothers come second, and fathers, at least in the cohort studied, had little influence. The study, published in the Journal of Personality and Social Psychology, followed 705 people and their families over three decades, starting in the 1990s.

British psychiatrist and psychoanalyst John Bowlby developed attachment theory in the 1970s and early ’80s, and it entered into the popular discourse in the intervening decades. The theory evolved, with subsequent research suggesting that our attachment styles are shaped across our lifetimes by multiple relationships, not just those with our parents, as Bowlby had initially proposed.

But until now, few studies had experimentally tested, over a person’s lifetime, the fundamental assumptions underlying attachment theory. To do this, Keely Dugan, an assistant professor of social personality psychology at the University of Missouri and her colleagues, analyzed data from one landmark longitudinal study of 1,364 children and their families that started in 1991 and stretched over 15 years. They then followed up with 705 of the original study participants, who were now 26 to 31 years of age.

ADVERTISEMENT

Nautilus Members enjoy an ad-free experience. Log in or Join now .

Read more: “Love Is Biological Bribery

The data for the original study came from a variety of sources: The authors periodically videotaped mothers and fathers interacting with their young children and made notes about their sensitivity to their children’s needs. They analyzed parent-child conflicts and closeness through reports from the parents and measured parents’ warmth and hostility through reports from the children. They also examined how the children rated their friendship quality and collected teacher and parent reports about their social competence with peers.

In the follow up, Dugan and her team evaluated the attachment styles and relationship quality of the now-adult participants, with their romantic partners, friends, and family members. They controlled for family income-to-needs ratio, maternal education, race and ethnicity, and sex assigned at birth.

Dugan and her colleagues found that a person’s relationship with their mother does shape their general attachment style and their specific individual relationships with friends, romantic partners, and fathers, accounting for 2 to 3 percent of differences in anxiety and avoidance. So, for example, people whose mothers were less warm and fuzzy during their younger years tended to feel more insecure in their adult relationships. The more recent the interaction with the mother, the more influence it potentially seemed to have. But early friendship bonds played an even bigger part than maternal relationships in the ways people navigated adult friendships and romantic partnerships, accounting for 4 percent of the variance in adults’ romantic partner- and best friend-specific attachment anxiety, and 10 to 11 percent in their partner- and best friend-specific avoidance.

ADVERTISEMENT

Nautilus Members enjoy an ad-free experience. Log in or Join now .

“In general, if you had high-quality friendships and felt connected to your friends in childhood, then you felt more secure in romantic relationships and friendships at age 30,” Dugani told Scientific American. “When you have those first friendships at school, that’s when you practice give-and-take dynamics,” she added. “Relationships in adulthood then mirror those dynamics.”

Even more reason to choose your schoolyard friends wisely.

Enjoying  Nautilus? Subscribe to our free newsletter.

Lead image: Ihnatovich Maryia / Shutterstock

ADVERTISEMENT

Nautilus Members enjoy an ad-free experience. Log in or Join now .

Read the original article

Comments

  • By taurath 2025-11-220:2010 reply

    No, one study doesn’t upend the last few decades of understanding of emotional attachment.

    The study simply says that ability to connect w friends is more predictive than observations they made of apparent attachment of parents.

    This happens much later so of course it’s more predictive of the actual end effects - that’s when attachment styles actually show up for the first time. Kids grow up to be very adaptive toward their parents but when they get to the rest of society that’s when the failures of connection and the failed bids for attention show up.

    A very resilient kid will do fine with friends even with a very bad attachment environment. A very sensitive kid or one with developmental problems will struggle in social environments.

    • By parpfish 2025-11-221:053 reply

      One study doesn’t definitively prove anything, but this is a 30 year longitudinal analysis with 700 participants. It’s way bigger than a typical study

      • By taurath 2025-11-225:443 reply

        The study itself doesn't say anything of the sort that the article title and this thread title do.

        I gaurantee you that if you polled any number of therapists what people's hangups are about it would be more likely to be the parents. Everyone I know is an inheritor of some significant amount of their family's generational trauma.

        • By pbhjpbhj 2025-11-2218:161 reply

          Tiny anecdote - it was only after being a parent for a decade or so that, when doing family history, I realised that my dad's parenting style -- which influenced my own so much -- was affected by being essentially fatherless; he had no direct model to base his own parenting on. WWII affecting children's upbringing ~70 years later.

          I think he did a great job when I was younger, but we haven't maintained a strong bond as adults (which is my fault as much as anything); something I want to try and change with my own children.

          • By bravura 2025-11-2220:12

            Call your father. And just chit chat for ten minutes.

        • By owenversteeg 2025-11-2219:15

          >if you polled any number of therapists what people's hangups are about it would be more likely to be the parents. Everyone I know is an inheritor of some significant amount of their family's generational trauma.

          I completely agree.

          The study looked at “individual differences in general attachment anxiety and avoidance in adulthood, as well as adults’ relationship-specific attachment orientations in each of their close relationships” - each quantified. Of course, the study isn’t about the source of trauma, but I don’t think you can quantify any of these things - the source of trauma, the source of anxiety, or the source of attachment styles.

          Imagine you have a very frugal parent. Maybe that early influence makes you a big spender. Maybe it makes you cheap. Maybe you spend normally, but you’re always anxious about not having enough. The same goes for other parental influence, good or bad. How does something like that show up on a chart? You can’t quantify it and you can’t generalize it.

          That said: I do think your early friendships and early relationships are full of useful hints. There are a lot of things caused by your parents that might be camouflaged (e.x. because the parents are older) that your own early friendships and relationships reveal.

        • By ajkjk 2025-11-2218:16

          I could see childhood friends mattering a lot precisely because it's the thing that can save you from the problems with your parents.

          Bad parents but good early friendships=turn out ok=don't show up in the data. Bad parents and bad early friendships = have it rough, show up in the data. The parents are the cause but the correlation to early friendships is even stronger because of the mediating effect.

      • By pfannkuchen 2025-11-225:02

        > ability to connect w friends is more predictive than observations they made of apparent attachment of parents

        So for comparing studies all measuring this^, yes that’s true. But there could be a flaw in the methodology here, where their observations of parents and interpretation thereof may not be predictive even while the totality of parent behavior is.

      • By Den_VR 2025-11-2210:39

        Can you link the study itself? What are the demographics of the participants like, the usual that’s clustered on one culture?

    • By scrubs 2025-11-221:381 reply

      I'm not a psychologist or psychiatrist. My observation is the more difficult cases of attachment in important adult relationships esp. partner/spouse is far more impacted by parents and their relationship than friends.

      This doesn't gain say that in the ages of 15 to say 35 peer interactions are not there or impactful to the worse or better but extremes in the nuclear family are not to be underestimated.

      • By hooskerdu 2025-11-224:381 reply

        It could be said that for any of us to think we could understand - with such a relatively short and still arguably shoddy understanding of the mind - or especially could say… is possibly insane.

        • By scrubs 2025-11-2210:21

          Not sure what you're after in that comment. The entire description of mind is maybe not shoddy but definitely high level functional meaning it always comes with a Chinese menu of if/and/but on a scale of emphasis in linear combination with other facets.

          Attachment modalities as far as I can see describes some human dynamics.

    • By jonahx 2025-11-221:16

      Also, even the proposed effect is modest:

      > But early friendship bonds played an even bigger part than maternal relationships in the ways people navigated adult friendships and romantic partnerships, accounting for 4 percent of the variance in adults’ romantic partner- and best friend-specific attachment anxiety, and 10 to 11 percent in their partner- and best friend-specific avoidance.

      Just slightly less modest that analogous parental predictors, according to their claims.

    • By popalchemist 2025-11-226:04

      Bingo. This kind of science reporting is the worst.

    • By aidenn0 2025-11-224:422 reply

      My own anecdotal experience matches this.

      I have a (now adult) child who was diagnosed with Reactive Attachment Disorder. She changed friends every 6 months, burning bridges behind her. She also cultivated the least-healthy friendships possible in whatever environment she found herself in.

      • By baconbrand 2025-11-225:001 reply

        How is she doing now?

        • By aidenn0 2025-11-2217:29

          Pretty well, all things considered.

      • By balamatom 2025-11-229:061 reply

        What was the etiology?

        • By aidenn0 2025-11-2217:181 reply

          Entered foster care at 2 due to abuse and neglect; we were her fourth placement in 7 months; weekly mandated visits with neglectful mother and monthly mandated visits with abusive grandmother for 2 years before termination of parental rights. She had nightmares until 3am after every visit with grandmother. We adopted her at 5.

          • By balamatom 2025-11-239:21

            This sounds terrible. Thank you for saving her.

    • By puppymaster 2025-11-221:55

      also given the psychology field research replication crisis, I would wait to see if this research can be replicated down the road.

    • By tootie 2025-11-2221:19

      I read the same conclusion in a NurtureShock published 15 years ago and they cited similar studies.

    • By standardUser 2025-11-2212:151 reply

      It kind of makes sense. We don't go out into the world thinking "ok, time to meet more my moms!". We can't treat anyone else like our mom, and vice versa. The mother-child relationship is probably the most important and influential for most people, but it's a complete anomaly. Friends are not an anomaly, their pretty typical, and since Mom and Dad are not friends, a young child has learned very little firsthand about friendship until it happens. Then they have some experiences with those friends and those experiences become better predictors of future attachment style than the dynamic with the mother. This makes sense because in the realm of friends, these are their formative experiences. Mom may have had an impact (and does according to the study), but the bigger influence on how you deal with future friends is in how you experienced your formative friends.

      • By em-bee 2025-11-2220:52

        i didn't start having friends until i entered university. and i spent my whole life searching for the kind of relationship that i wanted but didn't get from my parents. only a few years ago i realized that this is what i was searching for, and then i finally found it by realizing that i could be the parent to my kids that i had wanted my parents to be to me. my actual parents are more like friends to me than parents.

    • By cyanydeez 2025-11-2211:44

      That history: women are to blame

    • By hopelite 2025-11-2216:13

      Are you familiar with what the understanding is that differentiates/causes (very) sensitive vs (very) resilient kids, and whether and how that can be modulated in children or even the effects of it overcome in adults?

      I believe this is and will continue to be a growing issue as things like helicopter parenting and extremely anxious social dynamics, which were only exacerbated through the psychological abuse of the pandemic, starts becoming ever more dominant in society. I myself am a bit taken aback sometimes when taking to younger people and people who grew up in America specifically, and their expressed disbelief and exacerbation about totally normal things like, e.g., children walking to and from school or playing outdoors unsupervised until dusk.

      I don’t know if there is a short term solution outside of individual or small group choice as long as our societies are being ravaged by a ruling class that is making everything worse and more dangerous by the day, but I find the topic both intriguing as well as sad and foreboding.

  • By pedalpete 2025-11-2122:395 reply

    My first reaction was to refute this, but I think I've convinced myself this may be correct, assuming attachment styles are the right frame.

    I've been painted with the Avoidant brush, and logically it makes sense, broken home, removed from mother, moved regularly changing schools once a year for 5 years.

    However, my siblings are the opposite. We come from the same house, they didn't change schools as often as I did, which made me wonder how we could be so different.

    But when looked through the lens of friendships forming the attachment style, it makes more sense. I changed schools more often than my siblings, and therefore had more friendship changes, and less ability for attachment.

    • By jcims 2025-11-2123:393 reply

      Similar story here. Six schools by seventh grade. I think it does mess with you a bit.

      • By bsenftner 2025-11-2213:55

        Me too, 6 schools by 7th grade, with 3 city moves. I unfortunately had a stutter, which made forming friendships harder, I basically gravitated to the outcasts. And I read a shit ton; by the end of 4th grade I was finishing the entire classic section of the high school library. Reading way over my head, I joined an adult book club and had my first introductions to serious intellectual debates; which really put a distance between my same age peers and my interests. My stutter did not really get under control until my mid 20's.

      • By kimfc 2025-11-223:291 reply

        Yeah I'm the same, I think I went to nine schools by the time I went to college in the fall of 2019, most of the school changes happening in elementary school. It really does effect your ability to make friends

        • By pedalpete 2025-11-224:59

          I've found that I don't have trouble making friends, but I've put myself in situations where friendships come and go.

          I went from moving around a bunch, and making new friends at each place, to living in Whistler, BC, where you've got an annual turnover of new people, then I settled down in Bondi Beach, Australia, which doesn't have the turnover of Whistler, but not far off.

      • By faidit 2025-11-221:34

        Same. The only friends that stuck around were people from the internet.

    • By interroboink 2025-11-2123:301 reply

      Also, beware of taking generalities (such as the claims of this study) and applying that directly that to your specific life, or anyone else's.

      I mean, I like your comment and am glad you got thinking about this, but it's just a line of reasoning that I see a lot and I wish I saw less, so that's why I bring it up (:

      "True for most people" does not imply "true for me" or "true for that person over there".

      And the reverse is not valid either, of course - "true for me" does not imply "true for most people."

      There's always some tension between people's individual anecdotes and experiences (which are fascinating, and I like), and the claims of broader studies like this one.

      Sometimes I try to remind myself of this with the "on average, people have 2.3 children" factoid. Obviously, nobody actually has 2.3 children; the general truth does not necessarily apply to specific individuals; potentially not even a single one.

      • By pedalpete 2025-11-220:30

        100% agree. I actually think of attachment styles like this generally. Your upbringing does not dictate your life, it influences.

    • By cheesecompiler 2025-11-220:314 reply

      The family is a system, with different roles played by each participant. For instance, in toxic families, there is often one scapegoat, with an anxious attachment style, that affords the avoidant types in the family to participate in delusions.

      What are the dynamics like of everyone in your family?

      • By elbear 2025-11-228:362 reply

        I wanted to say the same: parents don't treat all children the same. For example, I have the feeling that the first child is the "practice" child. The parents learn from the mistakes made with them and don't repeat them with the children that follow. I don't know if there's any research to back this up and yes, I am a first born.

        • By toxik 2025-11-2210:39

          I think sequel kids, more than anything, benefit from having a trailblazer to refer to. It's no doubt true that parents get better at the job, but kids learn from demonstration. Older sibling is hypersensitive and has a hard time keeping friends around -> I better learn to swallow my pride. That kind of thing.

        • By thisislife2 2025-11-2210:26

          I've observed the same - unfortunately, first time parents are forced to try out all kinds of parenting experiments on their first-born, before they figure out how to be "good" parents. And subsequent kids, especially if they have them after some gap, get the benefit of this experience. Add to the woes of the first-born, they not only have to deal with normal sibling jealousy (of having to share their parents affection), but also resolve the emotional issue of why their younger siblings have an "easier" time (i.e. why their parents treat them "differently").

      • By pedalpete 2025-11-2223:40

        I am the scapegoat, though more commonly referred to as the black sheep.

        I am not anxious, I'm the avoidant one. I don't think any of my siblings are anxious either.

      • By softsound 2025-11-228:29

        Wow that explains a lot

      • By balamatom 2025-11-229:211 reply

        Remember how the modern "nucular" household is largely based on a modernization of the Roman patriarchal property distribution model, where the oldest male was ascribed the identities of all members of his household, and vice versa?

        That must've been extremely efficient for legal and accounting purposes, once. But, well, the only theory of mind anyone could develop in such circumstances involves grinding minds into fine paste. (There's a reason the Stoics are "seeing" an AI-driven resurgence, even though what'd be most appropriate for their target audience is probably again Skinner.)

        Remember how a great deal of how we live our "personal" lives was invented in a slaveholding state which mandated belief in gods and demons. And the rest in another.

        We are taught to consider all of this legacy cultural structure in terms of "haha how quaintly did people live 1000-2000-3000 years ago, were they stupid". Yet most of it lives on in some marginally altered form due to sheer global force of habit.

        Take Western human naming schemes for example: does your government permit you to change your name? do you inherit one or both granddads' names? do you get a patronym? extra personal names? are you also the security force for a place, like a Freiherr de So-and-So? and at what exact number of levels of recursive self-reflection does the word "person" stop meaning the role played, and starts meaning the human playing it?

        (When you're done with "identity", continue with "time-keeping" and begin to understand another psychological phenomenon causing much suffering - people's generalized inability to discern cause and effect.)

        The name - the sound through which individuals are conditioned to respond to the concepts of selfhood and identity (Foobert Barber Baznix! you come here right this instant! it is not me but you who is sleepy and hungry!) - is one of many such extremely arbitrary implementation details.

        Out of those emerges the thing sold to us by our caregivers and educators as "normal life" before we are able to know any better. That's the main way "primary socialization" has ever worked: a non-consensual intergenerational transmission of habits that have as much to do with self-soothing in the face of mortality as with practical concerns; in the end they just ascribe "imaginariness" to your memories of your mind being wiped, and the "you" is ready to go.

        Now, in the context of all those vague and admittedly entirely hypothetical "implementation details", proceed to imagine the troop of clothed primates not as a flat list of incidental blood relations, but as a dynamic system, a living group of conscious things; if you're feeling particularly scifi - a sort of distributed organism. What would be the purpose of the scapegoat organ in that organism? Do individual primates have an equivalent organ in their bodies? (Probably not the one you're thinking of but also a valid guess)

        • By imtringued 2025-11-2212:29

          There is no purpose of the scapegoat organ. This is one of the biggest fallacies people have with regards to natural selection and economics.

          Standard neoclassical economics theory tells people that they have perfect foresight and know the configuration/structure of all future possibilities. In other words, there are no unknown unknowns. You know everything you don't know yet.

          People have the same belief with regards to natural selection being efficient. It just seemingly chooses the most efficient organisms.

          In reality there is a developmental process with no guarantee of optimality or progress toward optimality. It is possible to get stuck in local maxima and it takes activation energy to get out of it.

          The scapegoat organ exists because the perceived marginal cost of fixing and investigating an incident or problem is considered more expensive than deflecting blame.

          The Iranians destroyed their water supply with scapegoats so trying to find a purpose in the scapegoat organ seems pretty insane. It's more like a weakness that leadership does not have a complete picture of the problems that its people are facing. You could argue that scapegoating is an expression of a lack of power. You have just enough power to blame others, but not enough to solve the problem.

  • By lordnacho 2025-11-2122:251 reply

    > But early friendship bonds played an even bigger part than maternal relationships in the ways people navigated adult friendships and romantic partnerships, accounting for 4 percent of the variance in adults’ romantic partner- and best friend-specific attachment anxiety, and 10 to 11 percent in their partner- and best friend-specific avoidance.

    Are those numbers r-squared figures? Seems like there's a lot more variance to be explained?

    • By dash2 2025-11-225:52

      Right. It also suggests two possibilities:

      1. Maybe the measurements are just very noisy. In which case they may also have other biases. 2. Maybe there are systematic causes which the study didn't capture. If so, controlling for them might change the results.

      Sigh. When I see a study headline like this I feel confident about two things. First, the study will have a weak design with no serious attention paid to causality, genetic confounding etc... second, the response to it will be full of people going "yes, that fits my N=1 anecdote" or "no that doesn't fit my N=1 anecdote", in other words, critiquing the weak methodology with an even weaker methodology (handwaving appeals to personal experience).

      One reason social science is hard is there isn't much market for the truth. People just want a nice story to tell themselves.

HackerNews