Skip to content

Attachment style08/27/202617 min read

How Adults Earn a Secure Attachment Style

Secure attachment can be earned in adulthood. What the research found, why anxious and avoidant leans need opposite practice, and how long it really takes.

Shadows of a dancing couple on a sunlit wall, he turns her under his arm
Contents
  1. What does secure attachment look like in adults?
  2. Can you become securely attached as an adult?
  3. The practice track if you lean anxious
  4. The practice track if you lean avoidant
  5. Do you need a partner to become secure?
  6. What does not work
  7. Common questions
  8. Sources

Marco's girlfriend has been quiet all evening. Two years ago he would have matched her silence, spent the night privately listing her flaws, and woken up half-convinced the relationship was a mistake. Tonight he asks one question, hears a tired story about her sister, and puts the kettle on. Nothing heroic happened. From the outside, a change of attachment style looks exactly this small.

Secure attachment is not a prize handed out in childhood and withheld from everyone else. Adults move toward it often enough that researchers gave the change a name, earned security, and the process resembles learning a second language. The first language, picked up at home before anyone could choose it, never disappears completely. But an adult can become genuinely fluent in a new one, fluent enough that on an ordinary Tuesday nobody hears the difference.

This guide goes through the research on earned security, then splits the practice into two tracks, because the anxious lean and the avoidant lean need opposite work. It ends with the part most advice skips, the real timeline and the strategies that change nothing.

Key points

  • Secure attachment in adults is a pair of learnable skills, asking for support and giving it, not a calm temperament. About half of adults already lean secure.
  • Earned security is documented, not motivational. Adults who reached security after a hard childhood function like the always-secure and mostly raise secure children.
  • Three factors repeat in the research on people who made the shift. One supportive figure, therapy lasting six months or more, and a stable partner.
  • Anxious and avoidant leans need opposite practices. The anxious track trains pausing and self-soothing; the avoidant track trains staying and disclosing.
  • The honest timeline. First shifts within months, depth over years, and relapses under stress are part of the process rather than proof it failed.

What does secure attachment look like in adults?

Secure attachment in adults is a working pair of skills. The person can ask for comfort without shame and give it without feeling swallowed. Roughly half of adults land closer to this corner than to any other, which makes the secure attachment style the most common of the four. It does not mean constant calm, an absence of jealousy, or never needing anyone.

The research version of that last point has a name, the dependency paradox. People with a reliable base take more risks, tolerate solitude better and recover faster than people bracing for abandonment or guarding their independence. Bowlby made the same argument decades earlier. Healthy self-reliance grows out of being able to lean on someone.

On an ordinary week, secure looks like this:

  • Saying "that stung" instead of staging a silence and waiting to be decoded.
  • Taking an evening alone without dressing it up as a disappearance.
  • Hearing "I need an hour" as information rather than a verdict.
  • Apologizing without treating the apology as a defeat.
  • Fighting, and then reaching out first anyway.

The last item carries the most weight. Secure couples do not fight less than everyone else; they repair sooner.

None of this requires having been born to perfectly attuned parents. Where the four styles come from, with a short self-check quiz, is in the guide to attachment styles. This article is about the road from insecure attachment toward this corner.

Can you become securely attached as an adult?

Yes. The evidence for earned security is some of the strongest good news in attachment research. Adults who grew up with cold, inconsistent or frightening care and function as secure anyway appear in study after study, and on most measures they are hard to tell apart from people who were secure all along.

The tool behind the finding is the Adult Attachment Interview. It does not score how good a childhood was. It scores how coherently a person tells the story of it. Adults who describe a hard childhood clearly, neither idealizing it nor drowning in it, get classified as secure, and their own children usually attach securely to them. Robert Karen, the historian of the field, put it in one line: "we are only doomed to repeat what has not been remembered, reflected upon, and worked through".

How do people get there? The Minnesota longitudinal study looked at mothers who had been abused as children and did not repeat the pattern with their own kids. Three factors kept surfacing: an alternative supportive figure in childhood, even a peripheral one; psychotherapy lasting six months or more; a stable, loving partner. None of the three is willpower. All three are relationships, and a pattern built inside relationships gets rebuilt the same way.

The change numbers hold both truths at once. Attachment style is about as stable as a personality trait, with a correlation near .50 across ten years of adulthood. Yet in one study about 30% of adults moved to a different attachment category within a few months, and a study spanning 59 years of adult life found attachment anxiety declining steadily with age. Sticky, and movable. Like an accent. Strong under stress, thinner with every year of practice.

The practice track if you lean anxious

The anxious pattern runs on hyperactivation. The alarm fires early, exaggerates the threat and demands action right now. Every practice that helps this lean does the same job. It puts space between the alarm and the action, long enough for the alarm to turn out wrong.

Wait twenty minutes before the protest move. The double text, the pointed silence, the jealousy test. Researchers call these protest behavior, and they survive because they occasionally work, the same schedule that keeps people at slot machines. This is a delay, not permanent suppression. Set a timer, leave the phone in another room, walk. Most alarms do not survive twenty minutes of walking.

Check the fear against the record. Write the fear in one sentence ("he's losing interest") and under it what actually happened this week. Read both. The anxious alarm quotes feelings as evidence; the page makes it show receipts.

Ask directly instead of testing. A test gives a partner a chance to fail. A direct sentence gives them a chance to answer. "My head has been telling me all day that something is off between us. Can you tell me where we are?" is uncomfortable to say, and it works far more often than a staged silence.

Keep plans that do not depend on the relationship weather. The Thursday run with a friend, the class, the standing dinner that stays in the calendar even when the relationship wobbles. Each kept plan is small proof that a gap in contact is survivable.

Where the alarm comes from, and why reassurance never seems to last, is unpacked in the anxious attachment guide.

The practice track if you lean avoidant

The avoidant pattern runs the opposite way, on deactivation. Distance restores calm, so needs get muted before they even register, and the muting itself quietly burns effort. The practices here reverse the anxious ones. Instead of learning to pause, this lean practices staying present and saying things out loud.

Say one true sentence a day. "I liked this weekend." "Work rattled me." Avoidant people disclose the least of any group and, tellingly, do not open up more when the other person does. Suppression is the engine of the pattern, so a small disclosure is a rep against the engine. One sentence. Daily.

Name the pause and set a return time. Needing space is legitimate. Vanishing is what does the damage. "I can't talk about this well right now. Give me an hour, I'll come back at nine" turns the same exit into something a partner can survive. The practice is the coming back at nine, not the leaving.

Stay one extra minute in conflict. When the urge to leave the room arrives, stay for one more minute, feet on the floor, and say what is actually happening. "I go quiet when I feel accused" does more repair than an hour of debate.

Notice when the flaw catalogue starts. The mental list of a partner's defects tends to appear right after closeness peaks, after the wonderful weekend or the talk about moving in together. The timing is the tell. It is a distance move wearing the costume of judgment. Levine's counter-practice is blunt. One note a night of something the partner did right.

Let someone do something for you. In story-completion studies, avoidant adults tell stories where the person in trouble finds relief without ever asking anyone for help. That missing step is the whole difference. Accept the lift to the airport. Say what the hard week was actually like.

Amir Levine, who co-wrote Attached, later compressed the secure job description into five words: consistent, available, responsive, reliable, predictable. Practicing them from the avoidant side, on the days when silence would be easier, is what the change is made of.

If both leans feel true at once, closeness wanted and feared on the same evening, the mix is called fearful-avoidant, and it moves best with a therapist alongside; the fearful-avoidant guide explains why. The full mechanics of the distance machinery are in avoidant attachment and dismissive-avoidant attachment.

Do you need a partner to become secure?

No. A stable partner is one of the three documented routes, not an entry requirement. What the change needs is one reliable other person, and a partner is only the most convenient candidate for the job.

With a partner, the relationship itself is the training ground. Sue Johnson boils a working bond down to three questions, A.R.E. Are you accessible, are you responsive, are you engaged. Repair is the skill worth drilling. Coming back after a bad evening, on purpose, sooner than pride suggests. Couples therapy built on attachment (EFT) reports that roughly 7 in 10 distressed couples recover, with the honest footnote that much of that research comes from the method's own founders.

Without a partner, two things carry the same weight. The first is friendship. Attachment is not reserved for romance; a friend who reliably answers at 2 a.m. is doing attachment work, and so is learning to be that friend. The second is therapy, and not as a consolation prize. Bowlby designed the therapist's role as a temporary secure base, "a trusted companion" beside whom old expectations can be safely re-examined, and measured attachment change over a course of therapy is documented.

There is also a lab-tested exercise missing from every list in the top search results: security priming. Participants spend a few minutes vividly recalling a person with whom they felt completely safe. The room, the voice, what being met like that felt like. Repeated over days, the exercise shifted how people saw themselves and their relationships, with effects outlasting the sessions. Two honest caveats: a single session fades fast, and for strongly insecure people the wrong memory can backfire, since in some studies thinking of a close person before a stressful task raised distress instead of lowering it. As a five-minute supplement to real contact, it has more experiments behind it than most exercises on the internet.

What does not work

Four popular strategies absorb effort and return almost nothing. They are worth naming because each one feels like progress from the inside.

Insight without practice. Reading about attachment, diagnosing exes, labelling parents. It feels like movement, but the pattern lives in reactions, not in knowledge. This is grammar study without conversation. The theory is flawless, and the first real sentence still comes out in the old language. The interviews behind earned security reward a reworked story, and stories get reworked by new experiences, not by new vocabulary.

Waiting for a secure partner to fix it. Secure partners really do steady the people they love; the buffering effect is documented. But as a plan it is passive, and the earned-security research lists a partner as one factor among three, working through what the person practices inside the relationship. Someone who outsources the whole job usually imports the old pattern into the new relationship within a year.

Playing unavailable. Replying slower on purpose, manufacturing distance to keep the upper hand. Whatever that wins tactically, mechanically it is avoidance practice, a rehearsal of the exact moves this whole project exists to retire. And if the games run inside a live pursue-withdraw loop with a real partner, the loop needs naming first; that dynamic has its own guide, the anxious-avoidant trap.

Speed-running. Thirty-day challenges promise a new attachment style by next month. The therapy that shows up in earned-security research ran six months or longer. First felt shifts arrive within months of consistent practice, depth takes years, and stress reliably brings the old accent back for an evening. A relapse after a hard week is not the project failing. It is what practicing under load looks like.

So the realistic promise reads less like a transformation ad and more like the thing adults do all the time. A second language learned late, spoken with a trace of accent, understood perfectly by the people who matter.

Common questions

First noticeable shifts typically arrive within a few months of consistent practice or therapy. The therapy documented in earned-security research lasted six months or more, and deeper change is measured in years. Relapses under stress are expected along the way and do not reset the clock.

Yes. The change requires one reliable other person, not necessarily a romantic partner. Close friendship, a therapist working as a temporary secure base, and structured recall of safe relationships (security priming) all have documented effects.

No, but it is the best-studied route. In research on people who broke a generational pattern, therapy of six months or more was one of three recurring factors; the other two, a supportive figure and a stable partner, are also relationships rather than techniques. Therapy is simply the version of a reliable relationship that can be scheduled.

Earned secure attachment is the research term for adults who grew up with difficult or inconsistent care yet function as secure, shown by their ability to tell the story of that childhood coherently. By most outcomes, including how their own children attach, earned-secure adults are hard to distinguish from people who were secure from the start.

Partly. A reliably responsive partner does turn the alarm down over time, and a stable partner is one of the three documented factors in earned security. The effect runs through repeated experiences inside the relationship, though, not through proximity alone. Treating a partner as a repair service usually bends the relationship back into the old shape.

Sources

  1. Bowlby, J. (1988). A Secure Base: Parent-Child Attachment and Healthy Human Development. Basic Books.
  2. Karen, R. (1998). Becoming Attached: First Relationships and How They Shape Our Capacity to Love. Oxford University Press.
  3. Mikulincer, M., & Shaver, P. R. (2016). Attachment in Adulthood: Structure, Dynamics, and Change (2nd ed.). Guilford Press.
  4. Carnelley, K. B., & Rowe, A. C. (2007). Repeated priming of attachment security influences later views of self and relationships. Personal Relationships, 14(2), 307–320.
  5. Chopik, W. J., Edelstein, R. S., & Grimm, K. J. (2019). Longitudinal changes in attachment orientation over a 59-year period. Journal of Personality and Social Psychology, 116(4), 598–611.
  6. Fraley, R. C. (2019). Attachment in adulthood: Recent developments, emerging debates, and future directions. Annual Review of Psychology, 70, 401–422.
  7. Johnson, S. (2008). Hold Me Tight: Seven Conversations for a Lifetime of Love. Little, Brown.
  8. Levine, A., & Heller, R. (2010). Attached: The New Science of Adult Attachment. TarcherPerigee.
  9. Taylor, P., Rietzschel, J., Danquah, A. N., & Berry, K. (2015). Changes in attachment representations during psychological therapy. Psychotherapy Research, 25(2), 222–238.
  10. Baldwin, M. W., & Fehr, B. (1995). On the instability of attachment style ratings. Personal Relationships, 2(3), 247–261.