Attachment style

What is an attachment style?

An attachment style (Dutch: hechtingsstijl) is the pattern by which someone seeks closeness or avoids it in close relationships. The idea comes from research into the bond between child and caregiver and was later applied to adult love relationships. It describes a tendency, not a personality type that is fixed.

The four styles

Three from the child research, and one added later.

1. Secure. Closeness feels within reach. You can lean on someone and be leaned on without it becoming fraught.
2. Avoidant. Distance feels safer than closeness. Self-reliance comes first, even where it isn’t needed.
3. Anxious-ambivalent. Closeness is wanted and uncertain at once. The attention goes to whether the other person will stay.
4. Disorganized. No consistent strategy: seeking and pushing away take turns.

Mary Ainsworth developed the Strange Situation Procedure to make those patterns visible in infants, and at first distinguished three. The fourth was added in 1983 by her colleague Mary Main, based on behavior that fit none of the three existing categories.

For adults a four-quadrant model is often used, proposed by Kim Bartholomew and Leonard Horowitz in 1991 in Journal of Personality and Social Psychology. There the avoidant side is called dismissive-avoidant or fearful-avoidant, depending on how someone thinks about themselves.

Where the theory comes from

From research on children, and only forty years later from research on lovers.

The theory was developed by psychiatrist and psychoanalyst John Bowlby. His starting point was that the bond with a caregiver is not a by-product of feeding but a biological system in its own right, one that organizes closeness when there is threat.

The step to adult relationships was taken by Cindy Hazan and Phillip Shaver in 1987, in Journal of Personality and Social Psychology. They understood romantic love as an attachment process and used the same three categories Ainsworth had found in children: secure, avoidant, and anxious-ambivalent. Their method was simple. Three short paragraphs, and adults picked which one best described how they felt and behaved in love relationships.

That is also the weakness of much popular use. The original measure was a self-description with three options, not a diagnosis.

What an attachment style means for a relationship

There is a link, it has been measured, and it works out differently for the two insecure styles.

Tianyuan Li and Darius Chan brought together 73 studies in European Journal of Social Psychology, with 118 independent samples and 21,602 participants. Their result: both anxiety and avoidance are linked to worse relationship quality, at the cognitive, emotional, and behavioral level.

The interesting part is the difference. Avoidance was more strongly negatively linked to overall satisfaction, connection, and support in the relationship. Anxiety was more strongly linked to more conflict. The effect of gender on those links was weak.

What that means in practice: someone with an avoidant tendency pulls back and the other person misses support; someone with an anxious tendency seeks contact and friction follows. Two very different problems, both under the same heading.

Why an attachment style isn’t fixed

Because the research on stability shows something more careful than the popular version.

Chris Fraley pulled together longitudinal data in Personality and Social Psychology Review and tested two models against each other: the prototype model, in which early representations keep playing a part, and the revisionist model, in which they are revised by new experiences.

His conclusion: “attachment security is moderately stable across the first 19 years of life and that patterns of stability are best accounted for by prototype dynamics”. Moderately stable, with a weighted stability of about r = 0.39 over intervals from one month to twenty-one years.

Moderate is not fixed. There is a lasting component to it, and there is room for change. Anyone who sticks a label on herself and uses it to explain why something can’t work is using the idea more strictly than the research allows.

How it shows up in bed

In two ways that feel opposite and share the same origin.

Li and Chan’s meta-analysis measures relationship quality and not sex, but the pattern they find is recognizable in the bedroom. In their data avoidance was more strongly linked to less connection and less felt support; anxiety more strongly to more conflict.

Put in terms of what people describe: with an avoidant tendency the need for closeness doesn’t disappear, it just gets awkward to show. Sex can then be easier than lying together afterward, and the conversation later feels heavier than the act itself. With an anxious tendency the attention goes to the other person instead. Is this going well, is this enough, will he stay. That makes it hard to notice what you want yourself, because there’s a question that has to be answered first.

What both have in common: the attention sits somewhere other than on your own experience. That is exactly where the sexual response cycle goes wrong, because arousal asks for attention that doesn’t need to go anywhere else.

These are averages from group research. They tell you something about patterns and nothing about you, and they’re not an excuse you can hide behind.

Why a label is so appealing

Because it puts order to something that feels confusing, and that is why it gets misused.

The four styles give language to something people otherwise have no words for. That’s a gain. Someone who reads for the first time that pulling back under stress is a pattern and not a character flaw recognizes herself and feels relief.

The downside is how easily such a word becomes an endpoint. A label explains nothing; it names. Fraley’s research also shows that the stability is moderate, so the pattern moves along with what you go through. Someone who says she’s just avoidant is using a description as an excuse.

There’s one more thing. People form different patterns with different people. So one label for a whole life doesn’t do justice to what has been measured.

What people get wrong about attachment styles

“You have one and it stays that way.” Fraley found moderate stability, not a fixed trait. The prototype model wins, but that means a lasting influence and not determinism.

“An online test tells you which one you have.” Hazan and Shaver’s original adult measure was three paragraphs you chose from. That’s a usable research instrument and not a diagnosis.

“Avoidant and anxious are two versions of the same problem.” The meta-research shows the opposite: avoidance is more strongly linked to less satisfaction and support, anxiety more strongly to more conflict.

“It says who you are.” It describes a pattern in close relationships. People can also have different patterns with different people, which further undercuts the urge to stick one label on someone.

“Avoidant means someone has no need for closeness.” The need is there; what’s missing is a way to show it. That difference is why pulling back so often reads as indifference, while under stress it can mean the exact opposite.

What you can do with this

• Use it as a description, not an explanation. “I pull back when it gets tense” helps; “I’m just avoidant” closes the conversation.
• Look at the pattern instead of the label. What happens with you when the other person comes closer, and what happens when they pull away.
• Count on it being able to shift. Moderate stability means experiences count.
• Don’t stick a label on the other person. The meta-research describes groups, and the effect of gender on those links turned out to be weak, which doesn’t support the usual assumption about anxious women and avoidant men.
• Ask more of a test. The original adult measure was three paragraphs you chose from; an online questionnaire that assigns you a label does something other than what the research did.
• Find a relationship therapist if the pattern is getting in your way. This page explains a concept and is not treatment.

More reading on attachment styles

Within this glossary, fight-or-flight is about what happens in a body under threat, which is related to the attachment reflex but not the same as it, and aftercare describes how people find each other again after an intense experience.

Outside this site, the meta-analysis by Li and Chan is the most useful if you want to know what was actually measured, and Fraley lays out how stable it really is.

Where this fits on this site

The word comes up in emails, usually as an explanation in advance: I’m avoidant, so don’t expect too much. You don’t have to do that. What does help is saying what you notice around closeness, because I can work with that. How a first conversation goes is covered under booking a date.

Sources

• Romantic love conceptualized as an attachment process. Cindy Hazan & Phillip Shaver, Journal of Personality and Social Psychology 52(3), 511–524, 1987. The article that brought attachment theory to adult love relationships. Source of the view of romantic love as an attachment process and the use of the three categories secure, avoidant, and anxious-ambivalent, taken from Ainsworth’s child research.
• Attachment Stability From Infancy to Adulthood: Meta-Analysis and Dynamic Modeling of Developmental Mechanisms. R. Chris Fraley, Personality and Social Psychology Review 6(2), 123–151, 2002 (DOI 10.1207/S15327957PSPR0602_03). A meta-analysis of longitudinal data testing the prototype model and the revisionist model against each other. Source of the conclusion “attachment security is moderately stable across the first 19 years of life and that patterns of stability are best accounted for by prototype dynamics”, and the weighted stability of around r = 0.39 over intervals from one month to twenty-one years. Read via the Crossref API, because the publisher blocks automated access.
• How anxious and avoidant attachment affect romantic relationship quality differently: A meta-analytic review. Tianyuan Li & Darius K.-S. Chan, European Journal of Social Psychology 42(4), 406–419, 2012 (DOI 10.1002/ejsp.1842). A meta-analysis of 73 studies with 118 independent samples and 21,602 participants. Source of the finding that both anxiety and avoidance are linked to worse cognitive, emotional, and behavioral relationship quality, that avoidance is more strongly negatively linked to overall satisfaction, connection, and support, that anxiety is more strongly linked to conflict, and that the moderating effect of gender was weak. The publisher’s site allows no scripts, so the abstract comes from Crossref.
• Attachment styles among young adults: A test of a four-category model. Kim Bartholomew & Leonard M. Horowitz, Journal of Personality and Social Psychology 61(2), 226–244, 1991 (DOI 10.1037/0022-3514.61.2.226). The article proposing the four-quadrant model for adults, with the distinction between dismissive-avoidant and fearful-avoidant. Only the bibliographic data were verified, via the Crossref API; the publisher deposited no abstract and blocks automated access, so no substantive quotation is drawn from this article on this page.
• Attachment theory. English-language encyclopedia. Source of the attribution of the theory to psychiatrist and psychoanalyst John Bowlby (1907–1990), the Strange Situation Procedure with which Mary Ainsworth identified attachment patterns, the initial three categories, and the fourth classification added in 1983 by Ainsworth’s colleague Mary Main based on behavior that fit none of the existing categories.

Related terms

An attachment style describes how someone organizes closeness; the terms below are about what happens in the moment itself.

• Fight-or-flight. The alarm response to threat, related to the attachment reflex but not the same as it.
• Aftercare. Finding each other again afterward, where attachment patterns become visible.
• Subdrop. The crash afterward, which for one person asks for closeness and for another for space.
• Consent. The agreement that doesn’t depend on what you’re used to asking for.
• Skin hunger. The longing for touch, which rises with anxious attachment and falls with avoidant.
• Relationship therapy. The form of treatment built on this model, with the numbers to back it.
• Pursuer-distancer pattern. The conflict behavior in which attachment anxiety becomes visible.
• Affair. Why avoidant attachment is most strongly linked to this, through commitment and not through lust.