Why we repeat the same relationship patterns

You’ve probably noticed the pattern. The same dynamic showing up in different relationships, with different people. You pursue someone who withdraws. Or you withdraw from someone who gets too close. Or, exhaustingly, you find yourself doing both – longing for closeness and feeling suffocated by it in turns. You tell yourself this relationship will be different. And then, at some point, it isn’t.

The frustrating thing is that understanding the pattern doesn’t seem to stop it. You might have read about attachment theory. You can probably name what you do. You might even be able to predict, in advance, the moment you’re going to do it. And then you do it anyway.

This isn’t a failure of willpower or intelligence. There’s a specific reason why insight alone doesn’t shift these patterns – and understanding that reason is, in my experience, the most useful place to start.

Where the pattern comes from

Attachment patterns don’t originate in adult relationships. They originate much earlier, before we have language for them or any way of choosing our response to them.

From the very beginning of life, the most important thing a child does is attach. Staying close to the people who provide safety is, for an infant, a matter of survival. And so the developing nervous system pays extraordinary attention to the caregiving environment: Is this person reliably available? Is emotional expression welcomed, or does it create distance? What does this person do when I need something?

From thousands of these interactions, the child builds what researchers call an internal working model – an unconscious template for what relationships are, what other people can be expected to offer, and what the self needs to do to keep the connection intact.

If the environment is consistent and warm, the child learns that connection is safe, that needs can be expressed, and that other people are generally trustworthy. This is secure attachment.

If availability is unpredictable, sometimes warm, sometimes withdrawn, the child learns to stay alert. To monitor the other person closely. To suppress their own needs when necessary to avoid disrupting the connection. This is the anxious pattern.

If emotional expression is consistently met with distance or discomfort, the child learns to self-contain. To manage internally. To keep relationships at a level of closeness that feels safe, which means: not too close. This is the avoidant pattern.

The template these experiences build doesn’t update automatically when the context changes. The strategy that protected attachment at nine years old continues operating at thirty-four, in a completely different environment, with different people, where the original situation no longer applies.

Why we’re drawn to what’s familiar

Here’s the part that I think most people haven’t heard clearly enough, because it explains something that otherwise makes very little sense.

The nervous system doesn’t seek what is emotionally healthy. It seeks what is emotionally familiar. Familiar means predictable. Predictable means manageable. And so, without any conscious intention, we tend to be drawn toward relationships that reproduce the emotional register we learned as normal, not because it’s good for us, but because it feels like home.

For someone with an anxious attachment pattern, that familiar register often includes a background hum of uncertainty about where they stand. The vigilance. The relief of reconnection. These are the textures of what love felt like growing up. As an adult, a relationship that produces those textures can feel intensely significant: real, meaningful, the right person. A relationship that is simply warm and consistent can feel, strangely, almost too easy. Not quite believable.

For someone with an avoidant pattern, genuine emotional intimacy, a partner who is open, expressive, and who needs them to be emotionally present, produces something that feels less like love and more like pressure. Not because they don’t want connection, but because their system learned that this kind of emotional closeness was something to be managed, not welcomed.

This is why the pattern repeats across different relationships. It’s not bad luck, and it’s not that you keep choosing the wrong people. It’s that your nervous system is doing what it was designed to do: orient you toward what it already knows.

Why understanding it isn’t enough

Attachment patterns don’t live primarily in conscious thought. They live in the nervous system, in the pre-conscious processes that generate emotional responses before rational thinking has had time to arrive. By the time you think ‘I’m doing the thing again’, the felt experience, the tightening in the chest, the impulse to withdraw, the urgent need to reach out, has already run its course.

This is why reading about attachment theory, as useful as it is, doesn’t on its own produce change. Understanding operates at a different level than the pattern. What shifts the pattern is sustained contact with it at the level where it actually operates: the body, the emotional responses, the core beliefs about what relationships are and what you deserve in them.

That kind of work is slower than gaining insight, and it’s more uncomfortable. It requires being willing to examine not just what you do in relationships, but what draws you to the ones you choose, and what makes security feel, at least at first, somehow less real than the familiar kind of longing.

What change actually looks like

The goal isn’t to eliminate the pattern – patterns encoded this early don’t simply disappear. What changes, through genuine psychological work, is the pattern’s authority. Its grip. The speed with which it runs before you’ve had a chance to notice. The automaticity of it.

Research on what’s called earned security, the process by which adults move from insecure to more secure attachment, consistently shows that this kind of change is genuinely possible. Adults who do this work are, on every measurable relational outcome, indistinguishable from those who have had security since childhood. The internal working model is not fixed. It updates, slowly and with the right kind of support, but it does update.

If you recognise yourself in any of what I’ve described here, that recognition is the beginning, not the solution, but the necessary first step. The question is what you do with it. Noticing the pattern without working with it changes very little. But noticing it, and then bringing the right kind of sustained attention to where it comes from and how it operates, can change a great deal. And in my experience, the people who do that work are rarely as stuck as they felt before they started.