Attachment styles in adults

Psychiatrist John Bowlby described attachment as an inborn system: when frightened, a person seeks closeness to whoever offers protection and feels safe near them (Bowlby, 1969). For toddlers, that’s usually their parents; in adults, a partner, close friends, and often parents too take on the role. The system switches on most strongly under stress, illness, or separation.

An attachment style describes the usual way someone runs this system. Some people find it easy to ask for support and just as easy to spend time apart; some fear being left; some prefer to cope alone. These habits grow out of experience, and psychology has studied them for more than half a century, since Bowlby.

A word about children: their attachment is studied with different methods, and the childhood categories don’t map onto adult styles one to one. From here on, this page is about adults and how the system of styles works.

Here are the attachment styles found in adults. Each has its own page with signs, causes, and how it changes, so here is just the gist. The literature sometimes uses other names, such as dismissing, preoccupied, or fearful, but they refer to the same four areas.

1. 1

   Secure. Both axes below average. Closeness isn’t frightening, time apart is handled calmly, and asking for support comes easily. More on the secure style page.

2. 2

   Anxious. Anxiety above average, avoidance below. The bond feels fragile, pauses are unsettling, and reassurance is needed again and again. How it works is explained in the anxious style overview.

3. 3

   Avoidant. Avoidance above average, anxiety below. Relying on yourself feels more natural than relying on someone else, and intense closeness prompts a step back. More about the avoidant style.

4. 4

   Fearful-avoidant. Both axes above average. The person reaches for closeness and fears it at the same time, hence the “closer, then farther” swing. Many call it disorganized, and the fearful-avoidant page explains what that really means.

For a long time, people chose their own style: they read a few descriptions and picked the closest one. In the late 1990s, researchers pooled dozens of questionnaires and found that nearly all of them measured two things: anxiety and avoidance (Brennan et al., 1998). The ECR-R questionnaire, which our test uses, is built on those two axes (Fraley et al., 2000).

Later studies asked whether styles exist as separate groups of people, and the answer was no: people are spread smoothly along the axes, without sharp borders (Fraley et al., 2015). That’s why styles are best understood as areas on a map. A point can sit deep inside an area or right at its edge, and those are different results. A point deep inside describes a pattern you’ll likely recognize at once; a point near the edge means two patterns blend.

Attachment styles weren’t invented overnight: the theory developed over about thirty years, from observing toddlers to questionnaires for adults. Each step kept what worked before and corrected what didn’t. Here are the key steps.

1. 1

   John Bowlby, 1969. Explained why a child needs attachment: an adult’s closeness protects, and the drive toward it is built into us as deeply as hunger. Bowlby didn’t yet describe styles; he described the system itself.

2. 2

   Mary Ainsworth, 1978. Described the Strange Situation experiments: a toddler is briefly left with a stranger, and researchers watch how the child greets the mother on her return. Her 1978 book set out three patterns: secure, avoidant, and anxious-ambivalent (Ainsworth et al., 1978).

3. 3

   Cindy Hazan and Phillip Shaver, 1987. Carried the idea over to adult love. Readers of a newspaper picked whichever of three short descriptions fit them best, and the breakdown looked much like the one seen in children (Hazan & Shaver, 1987).

4. 4

   Mary Main and Judith Solomon, 1990. Noticed that some children’s behavior fit none of the patterns: they moved toward their mother and then froze. That added a fourth childhood category, disorganized (Main & Solomon, 1990).

5. 5

   Kim Bartholomew, 1991. Split adult avoidance into two kinds: a calm kind, with a positive view of oneself, and a fearful kind, with self-doubt. That gave adults four styles (Bartholomew & Horowitz, 1991).

6. 6

   Scales instead of descriptions, 1998. Choosing a description gave way to measurement: first the ECR (Brennan et al., 1998), then the ECR-R, which measures each axis with 18 statements and compares scores with the norms of a large sample (Fraley et al., 2000).

Popular articles still often list three styles: secure, anxious, and avoidant. That’s the Hazan and Shaver scheme, carried over from childhood observations, and many websites use it to this day. But by the early 1990s it was clear that a single “avoidant” label hid two different experiences (Bartholomew, 1990).

For one person, distance comes easily: they feel sure of themselves and are used to doing without constant closeness. Another keeps distance out of fear: they want closeness but expect to be hurt in it. From the outside they look alike; from the inside they’re nearly opposites, and they need different kinds of help.

Two axes naturally create four areas, which is why the map has four styles. Fearful avoidance still isn’t the same as disorganized attachment, which is identified by observing toddlers or through a structured interview with adults; interview and questionnaire results barely overlap (Roisman et al., 2007). So our results use four styles, not three, and that’s also why each style gets its own page here.

Borders between areas run along the averages of Fraley’s norms. If a point lies very close to a border, within a quarter of the axis’s standard deviation, a single label would be a stretch, so the result also shows the neighboring styles. That happens for about 34% of people, so it’s not unusual.

A style can also differ between relationships. The first part of the test asks about romantic partners; the second asks, one by one, about the people closest to you overall, your mother, your father, and a friend. Attachment is often secure with one and anxious with another (Fraley et al., 2011). So a style is better understood as a description of how closeness feels to you now, not a lifelong trait. In six months, your point may be somewhere else.

A borderline result isn’t a test error. It says that both neighboring patterns are familiar to you, and it’s worth reading about each: people often recognize themselves most accurately in exactly that mix, more than in any single label.

1. 01Which attachment style is the most common?

   It depends on how you count. In surveys where people pick one of three descriptions, secure comes out on top: 59% in a US sample (Mickelson et al., 1997). When the two axes are split at the averages of Fraley’s norms, secure and fearful-avoidant come out roughly even, about a third each.
2. 02Which attachment style is best?

   Relationships tend to go more smoothly with a secure style: in research, anxiety and avoidance are linked to less satisfying relationships (Li & Chan, 2012). But that’s no verdict on a person. Each style once helped someone cope with their surroundings, each has strengths, and each can change.
3. 03Did Bowlby define the attachment styles?

   Bowlby himself didn’t sort people into styles: he described attachment as a system everyone has. Styles came later, with Mary Ainsworth’s observations of toddlers, and adult styles later still, with Hazan and Shaver and with Bartholomew. So “Bowlby’s four styles” is a misnomer: the four adult styles came later.
4. 04Is attachment style the same as personality?

   Personality traits describe a person in general, while attachment concerns close relationships specifically: the fear of losing someone and the readiness to lean on them. There are links to personality traits, but they’re moderate (Noftle & Shaver, 2006). So an outgoing extrovert can still turn out to have an avoidant style.