# Attachment Theory: The Four Styles and Where They Come From
Attachment theory is the idea that the way you were cared for as a child builds a working model of relationships — an expectation about whether people can be relied on — and that this model shapes how you handle closeness for the rest of your life. It began with John Bowlby in the 1950s, was made measurable by Mary Ainsworth in the 1970s, and was extended to adult romantic relationships by Cindy Hazan and Phillip Shaver in 1987.
It is one of the better-supported frameworks in psychology, and also one of the most distorted on the internet. The research describes tendencies measured on continuous scales; the popular version describes four personality types that explain everything about you. Those are not the same claim.
This page covers what attachment theory actually says, where the four styles come from, and what the framework can and cannot tell you about yourself.
What is attachment theory?
Bowlby's core proposal was that infants are born with a behavioural system whose job is to keep a caregiver close. Crying, clinging, and following are not learned habits but built-in strategies, because for most of human history proximity to an adult was the difference between surviving and not.
The consequential part is what happens next. Through thousands of small interactions — reaching out and being answered, or reaching out and not — a child builds what Bowlby called an internal working model: a set of expectations about whether people come when you need them, and whether you are the kind of person people come for.
That model becomes the default setting. It runs largely outside awareness, activates most strongly under stress, and shows up decades later in how you read a delayed text message or handle a partner needing space. Attachment theory's central claim is not that childhood determines your fate. It is that childhood installs a prediction, and predictions can be updated by evidence.
Where do the four attachment styles come from?
From an experiment. In the 1970s Mary Ainsworth designed the Strange Situation: a twenty-minute procedure in which a caregiver leaves an infant in an unfamiliar room and returns. What mattered was not the child's distress at the separation but their behaviour at the reunion.
Three patterns emerged. Some infants were upset, sought comfort on return, were soothed, and went back to playing — secure. Some were less visibly distressed and avoided the caregiver on return — avoidant. Some were highly distressed and could not be settled, alternating between clinging and resisting — anxious (originally ambivalent or resistant).
A fourth was added later. In 1990 Mary Main and Judith Solomon described infants who fit none of the three — freezing, approaching then backing away, or showing contradictory behaviours in sequence. They named this disorganized, because it lacked a coherent strategy rather than because the child was chaotic.
Hazan and Shaver's 1987 work found the same patterns in adult romantic relationships, which is where the framework you meet online comes from.
What are the four attachment styles?
Secure — closeness and independence both feel safe. Needs get stated directly, conflict ends in repair, a partner's separate life is not a threat. Roughly half to sixty percent of adults in Western samples.
Anxious (also preoccupied) — a strong desire for closeness paired with a persistent fear of being left. Vigilance about a partner's mood and availability, frequent need for reassurance, difficulty settling in ambiguity. Around one in five.
Avoidant (also dismissive) — independence is prioritised and closeness registers as a demand. Emotional distance under pressure, discomfort with dependence, care expressed through action more than words. Around a quarter.
Disorganized (also fearful-avoidant) — both patterns at once. Closeness is wanted and feared simultaneously, producing a push-pull that looks contradictory from outside. The least common, perhaps 5–15% depending on the measure.
Treat every percentage here as approximate. They vary by instrument and the samples are overwhelmingly Western and educated.
Are attachment styles really four types?
No, and this is the single most important correction to the popular version.
The research measures two continuous dimensions: attachment anxiety (how much you fear abandonment) and attachment avoidance (how uncomfortable you are with closeness). Every adult scores somewhere on both. The four styles are quadrants drawn on that two-dimensional space — low-low is secure, high-anxiety/low-avoidance is anxious, and so on.
That means there is no bright line. Someone scoring slightly above the midpoint on anxiety is not categorically different from someone slightly below it, though a quiz will hand them different labels. Most people cluster nearer the middle than the four vivid descriptions suggest.
It also means the honest question is not "which one am I?" but "where do I sit on these two dimensions, and does that explain something I have actually noticed?" The dimensions are what the evidence supports. The types are a communication device laid over them.
Can your attachment style change?
Yes. Attachment styles are stable but not fixed, and the term for acquiring security later is earned security — people who reach it are largely indistinguishable in outcome from those who grew up with it.
Change comes from sustained experience more than from insight. Knowing your style is useful for recognising a pattern; what actually updates the working model is repeated evidence that relationships can go differently — a consistent partner, a long friendship, an effective therapy.
One finding is worth singling out. The strongest predictor of earned security is not having had a good childhood. It is having a coherent narrative about the one you had — being able to tell the story, including the difficult parts, in a way that hangs together. The Adult Attachment Interview measures exactly this, and it predicts how someone's own children will attach better than the facts of their history do.
Which suggests something hopeful: what matters is less what happened than whether you have made sense of it.
What can't attachment theory tell you?
Quite a lot, and the popular version is silent on all of it.
It is not a personality theory. Attachment describes behaviour under relational stress. It says little about your ambition, humour, values, or how you work.
It is relationship-specific. Most people are more secure with some people than others. A single global label flattens that.
Self-report is weak. Online quizzes measure what you believe about yourself, and people under stress consistently read as more anxious than a longer measure would place them. The research instruments — the ECR and the Adult Attachment Interview — are more careful than anything you will take in five minutes.
It does not license diagnosis of other people. "My ex was avoidant" is usually a way of saying someone did not want what you wanted. Sometimes that is a pattern; often it is a preference.
And it does not explain everything. Attachment is one lens. It sits alongside temperament, culture, life stage, and what is currently happening in your life.
How attachment sits with other frameworks
Attachment style is most useful read against other things you know about yourself, because alone it over-explains.
Against the Big Five, attachment anxiety overlaps with neuroticism and avoidance with lower agreeableness — but only partially, and only in intimate contexts. Plenty of emotionally stable people have anxious attachment; plenty of warm people are avoidant with partners specifically. The trait model and the attachment model are describing different layers.
Against the five love languages, attachment often explains why a preference exists. Anxious attachment tends to pull toward words of affirmation and quality time, because both supply ongoing evidence of connection. Avoidant attachment tends toward acts of service — care through doing rather than declaring.
Against emotional intelligence, the two insecure styles produce mirror-image profiles: anxious attachment brings high awareness of others with lower regulation, avoidant brings solid regulation with less access to your own internal states.
Reading them together is more informative than any one alone — which is the point of holding more than one lens.
Common questions
Who created attachment theory? John Bowlby developed the theory from the 1950s onward. Mary Ainsworth made it empirical with the Strange Situation in the 1970s and identified the first three styles. Main and Solomon added disorganized in 1990. Hazan and Shaver extended it to adult romance in 1987.
Is attachment theory scientifically supported? It is among the better-evidenced frameworks in relationship psychology, with decades of longitudinal work behind it. The support is strongest for the two-dimensional structure and for attachment predicting relationship behaviour; it is weaker for the popular claims about fixed types and childhood determinism.
Can you have more than one attachment style? In practice, yes. Attachment varies by relationship, and most people show different patterns with a partner, a parent and a close friend. The single-label framing is a simplification.
What is the rarest attachment style? Disorganized or fearful-avoidant, at roughly 5–15% of adults depending on the measure used.
Does attachment style predict whether a relationship lasts? It predicts relationship satisfaction and conflict patterns reasonably well, and outcomes less well. A pairing of two insecure styles is not doomed, and two secure people are not guaranteed. Self-awareness and willingness to repair matter more than the labels.
Where to go next
If one of the four descriptions above sounded like you, each has a page of its own: secure, anxious, avoidant, and disorganized.
And if you want the wider picture — how relationships are actually going against work, health and everything else you are carrying — the Wheel of Life takes about ten minutes.
Your life is a story worth writing. Start your journal free — write a few lines whenever you like, and a warm letter comes back that reflects your own thinking and, over time, grows into a world of its own. No streaks, no blank page, nothing to keep up with. Free.
