Attachment Styles › Can You Change Your Style?
Can You Change Your Attachment Style? What the Long-Term Research Shows
Yes — attachment styles can change, and this is one of the better-evidenced findings in the field. A longitudinal study of more than 4,000 adults found that roughly a quarter of significant life events produced enduring changes in attachment security, not just temporary wobbles. A separate 59-year study (ages 13–72) found attachment anxiety declines on average across the lifespan, and avoidance decreases steadily too. Change is real, but it is neither fast nor linear: it happens through repeated experiences that contradict the old expectation, not through insight alone. Understanding your pattern is the start of the work, not the end of it.
This is the question underneath every other question people bring me about attachment. Once someone recognises their pattern, the very next thing they want to know is whether they are stuck with it. I want to answer that carefully — with the genuinely encouraging evidence, and also with the parts of the research that complicate the cheerful version you will find on most websites. You are better served by an honest picture than an inspiring one.
Sources: Fraley and colleagues (Journal of Personality and Social Psychology, 2021); Chopik and colleagues (Journal of Personality and Social Psychology, 2019); attachment stability synthesis (Scientific Reports, 2025).
Attachment does not change because you understood it. It changes because you lived something different, enough times, that your nervous system finally updated its expectations. — Dr. Prerna Kohli, PhD & M.Phil, Clinical Psychologist
What "Earned Secure Attachment" Means
The term comes from research in the 1990s describing people who, despite inadequate care in childhood, show secure attachment as adults — and who can talk about their difficult history coherently, without either dismissing it or being overwhelmed by it. That coherence turns out to matter enormously. Earned secures are not people who have erased their history; they are people who have integrated it.
This is the destination of the work described across this cluster: not becoming someone who never feels the old fear, but becoming someone for whom the old fear no longer has the final say.
How Change Actually Happens
Three mechanisms come up consistently in both the research and my clinical experience:
- Corrective relationships. The most powerful driver. Repeated experience of someone who stays steady — a partner, a close friend, a therapist — gradually contradicts the expectation that closeness ends badly. Being in a close romantic relationship has been found to predict lower attachment anxiety and avoidance over time.
- Self-awareness and a coherent narrative. Being able to see your pattern as it activates, and to tell the story of where it came from without collapsing into it or dismissing it.
- Repair, not perfection. What teaches the attachment system that rupture is survivable is not an absence of conflict — it is conflict reliably followed by reconnection. This is the mechanism people most often misunderstand.
Notice what is absent from that list: insight alone. In my experience plenty of people can describe their attachment history in impressive detail and still repeat the pattern exactly. Understanding creates the gap; repeated new experience fills it.
What Change Realistically Looks Like
Attachment researchers describe a "prototype" pattern: your attachment fluctuates week to week, but tends to pull back toward your characteristic level over time. In practice this means you will have stretches of feeling genuinely secure — open, trusting, unbothered by a slow reply — and then find yourself yanked back into old behaviour under stress.
I want to be emphatic here, because this is where people give up: the pull-back is not failure. It is what the process looks like. Progress shows up as secure stretches becoming more frequent, the pull-back becoming less intense, and — the marker I watch most closely — recovery time shortening. Someone who used to lose three days to an unanswered message and now loses twenty minutes has changed a great deal, even though the trigger still fires.
An honest complication, since you deserve the full picture
The earned-security literature is not entirely settled. A 23-year longitudinal study found that adults classified as "earned secure" had actual childhood caregiving that was comparable to that of continuously secure adults — suggesting that some of what looks like transformation may partly be a change in how people narrate their childhood, rather than a change in attachment's underlying structure. The same body of work found earned secures showed somewhat more depressive symptomatology than continuous secures.
Does that undo the case for change? No — the large longitudinal studies showing enduring shifts after life events stand on their own, and clinically I watch people change in ways that are hard to explain as mere re-narration. But it does mean the honest claim is "attachment patterns are meaningfully changeable," not "you can become a different person." Anyone promising the latter is selling something.
A short composite, drawn from patterns I see often — not a real client, and not a record of any one person.
Someone might come back to me a year into the work, disheartened, convinced they had made no progress at all — because the previous week their partner had gone quiet and the old panic had arrived exactly as it always did. What they had stopped noticing was everything around it. A year earlier that panic would have consumed a weekend and produced a dozen messages they regretted. This time they had recognised it within the hour, sat with it, and said something direct that evening instead. The trigger had not disappeared. Their relationship to it had changed completely. I have come to think that is the most common way progress gets missed: people watch for the fear to stop arriving, when what they should watch is how quickly they come back.
How Long Does It Take?
There is no honest single answer, and I am suspicious of anyone who gives one. What I can say from practice is that the direction of travel matters more than the pace, that anxious patterns often shift somewhat faster than avoidant ones (partly because avoidant people are less likely to seek support in the first place), and that trauma-rooted fearful-avoidant patterns need the most time and the gentlest handling. Most people notice meaningful change within months of focused work — not in how often the pattern fires, but in what happens next.
- Yes — roughly a quarter of major life events produce enduring attachment change (4,000+ adults).
- Attachment anxiety declines on average across the lifespan, and avoidance decreases steadily too.
- Change comes from corrective relationships, coherent self-narrative, and reliable repair — not insight alone.
- Expect fluctuation: progress means faster recovery, not the absence of triggers.
- The evidence has a genuine complication — some "earned security" may be narrative revision — so the honest claim is meaningful change, not transformation.
Want to actually do this work rather than read about it? Building earned security is among the most rewarding work I do, with individuals and couples. In-clinic in Gurugram and online across India and worldwide.
Book a Session WhatsApp Dr. Kohli