Avoidant Attachment: Why They Pull Away

They pull away because closeness sets off a threat response, not because they stopped caring. Avoidant attachment is a style built around self-reliance, and when a relationship starts to deepen, an old alarm goes off saying that depending on someone is risky. The nervous system responds by manufacturing distance. That is the entire mechanism, and once you can see it working, most of the confusing behaviour stops being confusing.

This guide covers what the research actually shows, how to spot the pattern, and what to do if you're the one on the receiving end of the withdrawal. Because the hard part usually isn't understanding avoidant attachment. It's knowing what to do at 11pm when they've gone quiet again.

What Is Avoidant Attachment?

Avoidant attachment, often called dismissive-avoidant attachment, is one of the insecure styles described in attachment theory. At the centre sits a high value on independence and real discomfort with depending on anyone, or having anyone depend on you. Closeness reads less like comfort and more like a loss of freedom.

People with this style frequently look like the most secure person in the room. Independent, capable, unbothered, good in a crisis. The pattern only shows up where intimacy does. When a relationship asks for vulnerability, ongoing reassurance, or emotional need, an internal system switches on to create room. Mario Mikulincer and Phillip Shaver, whose 2016 textbook Attachment in Adulthood is the standard reference in the field, call these moves deactivating strategies, because they shut the attachment system down rather than letting it stay switched on and connected.

There's a useful way to picture what's going on underneath. Kim Bartholomew and Leonard Horowitz, writing in the Journal of Personality and Social Psychology in 1991, mapped adult attachment onto two axes: how positively you view yourself, and how positively you view other people. Dismissive avoidance sits in the quadrant where the view of self stays positive and the view of others turns negative. That combination explains a lot. The self-sufficiency isn't a performance covering low self-esteem. They genuinely believe they're fine on their own, and they genuinely doubt that other people will come through.

How common is it? The benchmark figure comes from Mickelson, Kessler and Shaver's 1997 analysis of the National Comorbidity Survey, a nationally representative sample of American adults. They found roughly 59 percent secure, 25 percent avoidant, and 11 percent anxious. So about one in four adults leans avoidant. If you're dating one, you're not dealing with something rare.

And the share appears to be growing. A cross-temporal meta-analysis by Konrath and colleagues, published in Personality and Social Psychology Review in 2014, pooled 94 samples of American college students totalling 25,243 people between 1988 and 2011. Dismissing attachment rose from 11.93 percent to 18.62 percent over that window, while secure attachment fell from 48.98 percent to 41.62 percent. The increase held even after controlling for age, gender, race, and publication status.

Why Do Avoidant Partners Pull Away?

Nobody picks this pattern. It takes root early, in the relationship between a child and their caregivers, and it's a sensible adaptation to the situation the child was actually in.

When a caregiver consistently dismisses or minimises a child's emotional needs, the child adapts. Reaching out for comfort stops paying off, so the child stops reaching. Sometimes the message is blunt, a parent uncomfortable with tears, or one who praises a kid mainly for being tough and low-maintenance. Sometimes it's quieter, a home where achievement gets noticed and softer feelings don't. Either way the child draws a reasonable conclusion: depending on people doesn't work, so I'll depend on myself.

That conclusion is protective in childhood. The problem is it doesn't expire. The adult who learned early that needs were a burden can still feel, somewhere below thought, that leaning on a partner is unsafe, even when that partner is steady and kind.

Here's the part worth sitting with. Deactivating isn't the absence of feeling. It's the management of it. Under the withdrawal there's usually a person who wants closeness and doesn't trust it, which is a much more painful position than not caring would be.

The reward system runs quieter

There's brain imaging on this, and it's more specific than most people expect. Vrtička and colleagues at the University of Geneva scanned 16 volunteers while showing them faces paired with positive or negative social feedback, publishing the results in PLoS One in 2008. When a smiling face delivered approval, participants scoring higher on avoidant attachment showed markedly less activation in the ventral striatum and ventral tegmental area, the dopamine regions that normally light up for reward. The correlation was strong, at r = -.787.

Read that plainly and it says something quite sad. Warmth from another person registers as less rewarding. Not unnoticed, just flatter. So when your avoidant partner seems oddly unmoved by praise or affection you offered generously, you're probably not imagining the muted response.

One honest detail from the same study. Avoidant scores did not predict amygdala reactivity to angry faces. That threat response tracked with anxious attachment instead. Avoidance isn't a brain screaming danger at you. It's a brain that gets less of a lift from connection, which makes stepping back cost less than it would for most people.

What Are the Signs of an Avoidant Attachment Style?

Attachment sits on a spectrum and most people carry traces of more than one style. But a cluster of these usually points avoidant.

The need for space is non-negotiable

Alone time isn't a preference, it's a requirement. When a partner wants more togetherness than they have available, they feel crowded, sometimes trapped, and often resent the request without quite knowing why.

Closeness makes them notice your flaws

Just as things get serious, their attention drifts to everything wrong with you. The way you chew, your taste in films, a habit they used to find endearing. This is a classic deactivating move. It builds a case for the distance they already want.

Vulnerability gets intellectualised

Fears, needs, and tender feelings get analysed rather than felt. They'll explain an emotion rather than have it, change the subject, or handle something hard privately rather than let you in.

They withdraw right after real intimacy

A great weekend, an honest conversation, a genuinely close night, and then the retreat. This is the one that baffles partners most, because the pulling back arrives exactly when things were good. The closeness is the trigger.

Independence is a point of pride

There's real strength in self-reliance, and they know it. The cost only shows up when it becomes a wall.

There's some data on how this plays out. A 2023 study by Sagone and colleagues in the European Journal of Investigation in Health, Psychology and Education, covering 393 Italian adults aged 18 to 62, found that single participants scored significantly higher on discomfort with closeness than those in stable relationships. Each unit increase on that measure was associated with roughly 4.7 percent lower odds of being in a close relationship. You can read the full analysis via the National Institutes of Health archive. Discomfort with closeness has a measurable cost, and the people carrying it usually pay it quietly.

If some of this describes you rather than your partner, our guide to dismissive avoidant attachment goes deeper on the internal experience, and fearful avoidant attachment covers the version that mixes avoidance with anxiety.

Is It Avoidant Attachment or Do They Just Need Space?

Sometimes a person who wants two evenings alone a week is just a person who wants two evenings alone a week. The word avoidant gets thrown at anyone who isn't available on demand, and that's worth pushing back on, because mislabelling ordinary independence as pathology damages perfectly good relationships.

The base rates are a decent reality check. Bakermans-Kranenburg and van IJzendoorn's meta-analysis of the first 10,000 Adult Attachment Interviews, covering more than 200 studies over 25 years, put dismissing attachment at 23 percent of non-clinical adults, with 58 percent secure. So roughly one in four, not the majority. If everyone you've dated seems avoidant, the pattern might be in who you're picking, and our piece on why we pick the same partner is the more useful read.

Four questions separate a preference from a pattern.

• Does the distance follow closeness specifically? Someone who needs recovery time after a busy week needs it regardless of how the relationship is going. Avoidant withdrawal tracks intimacy. If the retreat reliably lands after the good weekend or the honest conversation, that's the tell.
• Do they come back and re-engage, or come back and pretend it didn't happen? A person who needed space returns and picks up where you left off. Deactivating usually involves the closeness being quietly unacknowledged afterwards, as though the conversation never took place.
• Can they tell you what they need? "I'm wiped, I need Sunday to myself, let's do Monday" is a request. Going quiet for four days and resurfacing without reference to it is something else. The difference isn't the space. It's whether you were told.
• What happens when you raise it kindly? This is the sharpest test. Bring it up once, warmly, with no accusation. Someone with a preference explains it. An avoidant pattern tends to produce deflection, a counter-criticism, or a sudden discovery that you're too needy.

One more thing. Plenty of behaviour that looks avoidant is really just a mismatch in how much togetherness two people want. That's a compatibility question, not a diagnosis, and it deserves an honest conversation rather than a label. Working out how to communicate what you actually need will get you further than deciding which attachment style they are.

Does the Withdrawal Actually Damage the Relationship?

Yes, and the damage is mostly indirect. The withdrawal itself isn't what wrecks things. It's the loop it starts with the other person.

The cleanest data on this comes from Bretaña and colleagues, who studied 175 heterosexual Spanish couples aged 18 to 72 and published in Frontiers in Psychology in 2022. Because they measured both partners, they could trace what each person's behaviour did to the other. Avoidance strongly predicted withdrawing during conflict, identically for men and women at a standardised coefficient of 0.41. Withdrawal was by far the avoidant person's signature move, more so than getting demanding or aggressive.

Then the loop. One partner's withdrawal predicted the other partner's demanding and aggressive response, at 0.46 when men withdrew and 0.48 when women did. So the retreat doesn't produce calm. It produces escalation from the person left standing there, which produces more retreat.

The satisfaction numbers are blunt. Avoidance correlated with relationship satisfaction at r = -0.63 for both men and women, which is a large effect by any standard in this field. And the withdrawal was doing the work: it statistically mediated the path from avoidant attachment to the person's own dissatisfaction, and it reached across to lower their partner's satisfaction too, by way of the demanding response it provoked.

Two things follow from that, and they cut in opposite directions. If you're the avoidant one, the distance you use to protect the relationship is measurably corroding it, including for you. And if you're the partner, the escalation that feels completely justified is a documented part of the machinery, which means your side of it is also something you can change. Nobody in this pattern is only a bystander.

Why Does Giving Them More Love Backfire?

Because being singled out for extra attention threatens an avoidant person more than being left out does. That sounds backwards. It's also the single most counterintuitive finding in this area, and it explains why the instinct to fix distance with more affection reliably makes things worse.

Izaki, Wang and Kawamoto tested this directly and published it in Frontiers in Psychology in 2022. They put 133 students through a ball-tossing game with three setups. In one, you get frozen out and receive 2 throws out of 30. In another, you get a fair share at 10 of 30. In the third, you get showered with 15 of 30, far more than anyone else. Afterwards everyone rated how much the experience dented their sense of belonging, self-esteem, control, and meaning.

The exclusion results went as you'd expect from everything else in this guide. People high in avoidance took less of a hit to self-esteem than people low in avoidance when they were left out, at p = 0.02. Belonging, control, and meaning showed no real difference. Being ignored simply costs them less.

The over-inclusion results are the interesting half. When high-avoidance participants were given more attention than everyone else, they reported significantly greater threat to both belonging and self-esteem than low-avoidance participants did. They also felt more threatened by the extra attention than they had by an equal share of it. Getting more was worse than getting the same. And anxious attachment produced no significant effect in any condition, so this isn't a general insecurity thing. It's specific to avoidance.

Sit with what that means at home. The extra affection you offer when they've gone quiet, the surprise weekend, the long message about how much they mean to you, the intense burst of closeness meant to close the gap, all of it is over-inclusion. You're not giving them a reason to come closer. You're increasing the exact input their system reads as threatening, which is why the response so often looks like ingratitude.

So the practical rule isn't less love. It's steadier love, at a level that doesn't spike.

• Aim for equal footing, not extra. The equal-inclusion condition was the comfortable one. Treating them like a normal participant in your life beats treating them like a project.
• Spread it out instead of concentrating it. Six small ordinary contacts across a fortnight land better than one enormous gesture, which is just a spike with a bow on it.
• Watch your own escalation. The urge to give more when someone withdraws is human and it's the wrong direction. Distance is not a signal to increase the dose.
• Don't read the flat response as rejection of you. A muted reaction to a big gesture is the pattern doing what it does, not a verdict on your worth.

Worth keeping this in proportion. It's 133 undergraduates tossing a virtual ball, not couples in their kitchens, so treat the direction as instructive rather than settled. But it lines up neatly with the softening research below, and with why intense early attention can feel suffocating rather than flattering. If the over-attention in your relationship is running the other way and feels strategic rather than clumsy, our guide to love bombing covers where that line sits.

How Do You Deal With an Avoidant Partner?

Most advice here is either wishful or punitive. What follows is neither. It works because it changes the input their nervous system is responding to.

The strongest evidence for any of it comes from Nickola Overall, Jeffry Simpson and Helena Struthers, who video-recorded 180 heterosexual couples working through conflicts where one partner wanted the other to change something. Their findings, published in the Journal of Personality and Social Psychology in 2013 and summarised in this NIH-archived review of partner buffering, were clear. Avoidant partners on the receiving end of a request showed more anger and more withdrawal, and those discussions went worse. But at the specific moments when their partner used what the researchers called softening communication, reassurance of the person's value and acknowledgement of their autonomy, the anger dropped and the withdrawal didn't arrive. Same avoidant person, same conflict, different outcome depending on how the request was delivered.

That's the practical heart of this. You can't argue someone out of avoidance, but you can change the conditions that trigger the defence.

1. Stop chasing during the withdrawal. This is counterintuitive and it's the highest-leverage change you can make. Pursuit reads as pressure, pressure triggers more deactivating, and you end up in the push-pull loop described in our guide to the anxious avoidant trap. Stepping back isn't a tactic to make them miss you. It removes the thing that's making them run.
2. Say what you need without wrapping it in an accusation. There's a real difference between "you always shut me out" and "I feel disconnected and I'd like an evening together this week." The first is a verdict and invites defence. The second is a request and can actually be answered.
3. Be boringly predictable. Avoidant nervous systems are calibrated for inconsistency. Doing what you said you'd do, showing up when you said you would, and reacting proportionately when things go wrong slowly builds a different expectation. It's unglamorous and it's the main thing that works.
4. Give space with a sentence attached. Silent withdrawal on your side confirms their model that closeness is unstable. Saying you're around when they're ready gives them room without leaving them guessing. Then actually leave them alone.
5. Notice the direction of travel, not the good days. Everyone has good weeks. What matters is whether the gaps are getting shorter over months and whether they're willing to name the pattern themselves. If nothing is moving after a year of your effort and none of theirs, that's information.
6. Keep your own life intact. Organising your entire week around someone's availability is how resentment gets built. It also makes you less steady, which makes them more avoidant. Your friendships and your own plans aren't a consolation prize.

One boundary worth stating plainly. You can't do this work on someone's behalf. You can make the conditions safer, and that genuinely helps, but the actual change has to be something they choose. If you find yourself managing their emotions while your own needs go unspoken for months, the pattern has shifted into something closer to loving an emotionally unavailable partner, and that's worth looking at honestly.

What Do You Do When They Shut Down Mid-Conversation?

Stop the conversation. Not the relationship, not the topic, just the conversation, and set a specific time to come back to it. Nothing useful gets decided after someone has gone flat, and pushing through is the single most common mistake partners make.

Here's the part that changes how you read the moment. The blank calm on their face isn't calm. Work reviewed in Frontiers in Human Neuroscience in 2015, drawing together psychophysiological studies of attachment interviews, describes a consistent finding: adults using deactivating strategies show heightened skin conductance, a marker of sympathetic nervous system arousal, while reporting and displaying little distress. The researchers describe it as heightened physiological but dampened behavioural reactions.

So the person who looks unbothered while you're upset is, measurably, not unbothered. Their body is running a stress response and their expression is running suppression. That gap is the whole thing. And it means the flat face you're reading as indifference, or worse as contempt, is closer to overload.

What to do with that in the moment:

1. Name the state, not the character. "I think we've both hit a wall" works. "You're doing the thing where you check out" is a diagnosis delivered mid-argument, and nobody has ever responded well to one.
2. Put a time on the pause. An open-ended break feels like abandonment to you and gets quietly extended by them. "Let's pick this up after dinner" gives you both something concrete. Twenty minutes is usually the minimum for a nervous system to settle.
3. Agree the re-entry rule while things are calm. Decide in advance, on a good day, that either of you can call a pause and that whoever calls it is responsible for restarting the conversation. That last clause matters. Without it, the pause becomes the ending.
4. Don't chase resolution the same night. The urge to fix it before sleep is understandable and it usually costs you. A conversation resumed calmly on Tuesday beats one forced through at midnight.
5. Watch what happens after, not during. Anyone can shut down under pressure. The question is whether they come back. Consistent returning is the signal worth tracking.

If pauses in your relationship reliably turn into days of silence, that's a different problem, and our guide to healthy conflict resolution covers how to rebuild a repair process from scratch.

Can Avoidant Attachment Change?

Yes, with an honest caveat. Attachment styles are learned, which means they can be relearned, and psychologists use the phrase earned secure attachment for people who grew up insecure and developed the capacity for steady closeness later. Research on how attachment patterns shift toward security is summarised in the NIH literature.

The caveat comes from the therapy evidence. Burgess Moser and colleagues tracked 32 couples through Emotionally Focused Couple Therapy and published the results in the Journal of Marital and Family Therapy in 2016. Couples significantly decreased in relationship-specific attachment avoidance across sessions, and attachment behaviour moved measurably toward security. But the two-year follow-up study from the same research group found those gains in avoidance weren't fully maintained over the long run.

That's not a reason to skip therapy. It's a reason to treat avoidant patterns as something you keep practising rather than something you fix once. The people who hold onto the change are generally the ones who keep doing the work after the sessions stop. If waiting lists or the idea of sitting in a room with a stranger are what's stopping you, online therapy with an attachment-trained counsellor lowers the barrier considerably, and for avoidant patterns specifically the option to message between sessions often suits people who process privately before they speak.

What actually shifts the pattern

Catching the deactivating in real time is where it starts. The next time the urge to retreat arrives, pausing to ask what just happened, and whether things got close or someone needed something, opens a window between the impulse and the exit.

Then naming it out loud instead of acting on it. Telling a partner "I notice I want to pull back right now, and I'm working on staying" is a small act of vulnerability that turns a confusing disappearance into an open door. It's uncomfortable and it's disproportionately effective.

After that it's repetition. One honest share, one request for support, one conversation you stay in a few minutes longer than is comfortable. Every time closeness doesn't lead to engulfment or rejection, the nervous system collects evidence that this might be survivable. Our secure attachment guide covers what you're building toward.

What Else Do People Ask About Avoidant Attachment?

Do avoidant partners ever come back?

Often yes, because distance lowers the pressure that triggered the withdrawal in the first place. But coming back isn't the same as change. If nothing has shifted in how they handle closeness, the same cycle usually restarts within a few months. Watch what they do differently, not whether they return.

Should I give an avoidant partner space or reach out?

Give space, but say something first. Silent withdrawal on your side reads as punishment and confirms their belief that closeness is unstable. A short message saying you're around when they're ready gives them room without leaving them to guess. Then genuinely leave them alone until they respond.

Is avoidant attachment the same as being emotionally unavailable?

They overlap but aren't identical. Avoidant attachment is a specific pattern with a known developmental origin and a known mechanism. Emotional unavailability is a broader description that can also come from grief, burnout, an existing relationship, or simply not being interested. The behaviour can look the same from outside.

Can two avoidant people have a healthy relationship?

Yes, and it's often calmer than an anxious and avoidant pairing because neither person is chasing. The risk is different. Two avoidants can drift into a comfortable but thin relationship where nothing difficult ever gets raised, and the distance grows slowly enough that neither notices until it's wide.

How long does it take for an avoidant partner to open up?

There's no set timeline, and it depends far more on repetition than on months passing. What builds trust is a pattern of small disclosures that are met calmly, over and over. Most people describe meaningful change taking a year or more, and it moves faster when the avoidant person is actively working on it.

If you recognise yourself in the avoidant side of this, be gentle about it. The pattern was a clever solution to a real situation and it kept a younger version of you safe. You're allowed to thank it and then let it loosen. Independence and genuine intimacy aren't opposites, whatever your nervous system currently believes.

Sources: Mickelson, K. D., Kessler, R. C., and Shaver, P. R. (1997). Adult attachment in a nationally representative sample. Journal of Personality and Social Psychology, 73(5), 1092 to 1106.  ·  Konrath, S. H., Chopik, W. J., Hsing, C. K., and O'Brien, E. (2014). Changes in adult attachment styles in American college students over time: A meta-analysis. Personality and Social Psychology Review, 18(4), 326 to 348.  ·  Burgess Moser, M., Johnson, S. M., Dalgleish, T. L., Lafontaine, M., Wiebe, S. A., and Tasca, G. A. (2016). Changes in relationship-specific attachment in Emotionally Focused Couple Therapy. Journal of Marital and Family Therapy, 42(2), 231 to 245.  ·  Sagone, E., Commodari, E., Indiana, M. L., and La Rosa, V. L. (2023). Exploring the association between attachment style, psychological well-being, and relationship status in young adults and adults. European Journal of Investigation in Health, Psychology and Education, 13(3).  ·  Vrtička, P., Andersson, F., Grandjean, D., Sander, D., and Vuilleumier, P. (2008). Individual attachment style modulates human amygdala and striatum activation during social appraisal. PLoS One, 3(8), e2868.  ·  Overall, N. C., Simpson, J. A., and Struthers, H. (2013). Buffering attachment-related avoidance: Softening emotional and behavioral defenses during conflict discussions. Journal of Personality and Social Psychology, 104(5), 854 to 871.  ·  Bretaña, I., Alonso-Arbiol, I., Recio, P., and Molero, F. (2022). Avoidant attachment, withdrawal-aggression conflict pattern, and relationship satisfaction: A mediational dyadic model. Frontiers in Psychology, 12, 794942.  ·  Mikulincer, M., and Shaver, P. R. (2016). Attachment in Adulthood: Structure, Dynamics, and Change (2nd ed.). Guilford Press.  ·  Bakermans-Kranenburg, M. J., and van IJzendoorn, M. H. (2009). The first 10,000 Adult Attachment Interviews: Distributions of adult attachment representations in clinical and non-clinical groups. Attachment and Human Development, 11(3), 223 to 263.  ·  Gander, M., and Buchheim, A. (2015). Attachment classification, psychophysiology and frontal EEG asymmetry across the lifespan: A review. Frontiers in Human Neuroscience, 9, 79.  ·  Izaki, R., Wang, Y., and Kawamoto, T. (2022). Avoidant attachment attenuates the need-threat for social exclusion but induces the threat for over-inclusion. Frontiers in Psychology, 13, 881863.