Attachment Repair in Adulthood: How to Rewire the Patterns Your Childhood Built Without Spending Years in Therapy

RELATIONAL TRAUMA & RECOVERY • April 2026

LAST UPDATED: APRIL 2026

The Sunday Night Feeling She Can’t Name

It’s 9:47 on a Sunday night. Maya is a 41-year-old pediatric surgeon — she has performed over two thousand procedures, trained residents, published in peer-reviewed journals, and navigated a hospital system that was not built with her in mind. She is, by every external measure, someone who has figured it out. Her partner is asleep in the next room. He is kind. He asks her how she’s doing. He notices when she’s tired. And yet Maya is sitting at the kitchen table with a glass of water she hasn’t touched, feeling something she doesn’t have a word for.

It isn’t unhappiness, exactly. It isn’t anxiety — she knows what that feels like. It’s more like a glass wall. Her partner is right there, and she can see him, and she knows he loves her, and she cannot let him in. She hasn’t told him she’s scared about the malpractice inquiry at work. She hasn’t told anyone. She’s been telling herself it’s because she doesn’t want to worry him. But sitting here in the quiet, she knows that’s not the whole truth. The whole truth is that she doesn’t know how to let someone see her scared. She’s not sure she ever learned.

This is what insecure attachment looks like in a high-functioning adult. Not dramatic dysfunction. Not obvious avoidance. A glass wall. A competence that has become a container. A woman who is brilliant at managing everything except the particular vulnerability of being known.

In my work with clients, I see this pattern constantly. The driven woman who has built an extraordinary life and still feels fundamentally alone in it. The woman who can lead a team of forty but can’t tell her partner she’s afraid. The woman who has read every book about attachment — John Bowlby, Mary Ainsworth, Sue Johnson — and understands her dismissive attachment style intellectually, completely, and still finds herself on the other side of that glass wall every Sunday night.

Understanding your attachment patterns is not the same as changing them. That’s the gap this article is about. Not the theory — you may already know the theory. The gap between knowing and actually being able to do something different in the moments that matter.

Attachment repair is possible. It’s not easy, and it’s not fast, and it doesn’t happen through intellectual understanding alone. But it is genuinely, neurobiologically possible. And the path toward it is more specific — and more accessible — than most people realize.

What Is Attachment Repair?

Attachment repair is the process of updating those internal working models — of giving the nervous system new relational experiences that contradict the old predictions. It’s not about erasing your history. It’s about adding new data to a system that has been operating on outdated information.

John Bowlby, MD, psychiatrist and psychoanalyst, developed attachment theory in the 1950s and 1960s, drawing on his observations of children separated from their caregivers during World War II. His central insight was that the need for attachment — for a reliable, responsive caregiver — is not a secondary drive derived from hunger or comfort, but a primary biological need. Infants who don’t have their attachment needs met don’t just feel sad. Their nervous systems organize around that absence. They develop strategies — avoidance, hyperactivation, disorganization — to manage the unbearable reality of needing someone who isn’t reliably there.

Mary Ainsworth, PhD, developmental psychologist, later identified the specific attachment patterns that emerge from different caregiving environments through her landmark Strange Situation studies. Secure attachment develops when caregivers are consistently responsive. Anxious-preoccupied attachment develops when caregiving is inconsistent — sometimes warm, sometimes unavailable — leaving the child in a state of chronic hyperactivation of the attachment system. Dismissive-avoidant attachment develops when emotional needs are consistently minimized or ignored, leading the child to deactivate the attachment system as a survival strategy. Disorganized attachment develops in the most frightening caregiving environments, where the caregiver is simultaneously the source of fear and the only available source of comfort.

Kim Bartholomew, PhD, psychologist at Simon Fraser University, later extended Ainsworth’s model to adults, identifying four adult attachment styles that map onto Ainsworth’s infant categories: secure, preoccupied, dismissive, and fearful. Her research demonstrated that these patterns are stable across adult relationships — not because they’re fixed, but because we unconsciously recreate the relational conditions that feel familiar.

Attachment repair, then, is the deliberate process of interrupting that recreation. It requires new relational experiences — experiences that are different enough from the old predictions to begin updating the internal working model. This is why trauma-informed therapy is so central to attachment repair: the therapeutic relationship itself is a new relational experience. A therapist who is consistently present, attuned, and non-abandoning provides the nervous system with data it has never had before.

But therapy isn’t the only path. Secure relationships — with partners, friends, mentors, even communities — can also provide the corrective relational experiences that update internal working models. The key is that the experience has to be felt, not just understood. The nervous system learns through experience, not through insight.

The Neurobiology of Attachment Repair

The neuroscience of attachment repair begins with neuroplasticity — the brain’s lifelong capacity to form new neural connections in response to new experiences. For decades, the dominant model in neuroscience held that the brain’s architecture was essentially fixed after early childhood. We now know this is wrong. The brain remains plastic throughout the lifespan, and the neural circuits that govern attachment — the circuits of the prefrontal cortex, the limbic system, and the autonomic nervous system — can be updated through new experience.

Stephen Porges, PhD, professor of psychiatry at Indiana University School of Medicine and developer of Polyvagal Theory, has contributed one of the most important neurobiological frameworks for understanding how attachment repair actually works. Porges’ theory describes three hierarchical states of the autonomic nervous system: the ventral vagal state (safety and social engagement), the sympathetic state (mobilization — fight or flight), and the dorsal vagal state (immobilization — freeze and shutdown). For individuals with insecure attachment, the social engagement system — the ventral vagal circuit — is often chronically underactivated. They’ve learned, at a nervous system level, that closeness isn’t safe.

Deb Dana, LCSW, clinical social worker and author of The Polyvagal Theory in Therapy, has translated Porges’ framework into clinical practice in ways that are directly relevant to attachment repair. Dana describes the process of nervous system regulation as fundamentally relational: we regulate our nervous systems through contact with other regulated nervous systems. This is what she calls co-regulation — the biological mechanism by which attachment relationships work. When a securely attached caregiver soothes a distressed infant, they’re not just providing comfort. They’re literally lending their regulated nervous system to the infant’s dysregulated one, helping the infant’s system find its way back to baseline.

For adults with insecure attachment, co-regulation is often the missing ingredient. They’ve learned to self-regulate — sometimes brilliantly — but they’ve never learned to let another person’s regulated presence actually land in their body. The glass wall Maya feels isn’t a character flaw. It’s a nervous system that learned, very early, that needing someone was dangerous. The ventral vagal circuit — the circuit of social engagement and co-regulation — went offline as a protective measure. Attachment repair, neurobiologically, is the process of bringing it back online.

Peter Levine, PhD, psychologist and developer of Somatic Experiencing, adds another dimension to this picture. Levine’s work, detailed in Waking the Tiger: Healing Trauma, emphasizes the concept of pendulation — the natural oscillation between states of activation and settling in a healthy nervous system. For individuals with insecure attachment, this pendulation is often disrupted. They’re either braced against connection (dismissive) or flooded by it (anxious). Attachment repair involves restoring the capacity to pendulate — to move toward closeness and then back to self, without either collapsing into merger or retreating behind the glass wall.

Daniel Siegel’s concept of integration is also essential here. Siegel, in The Developing Mind, describes integration as the linkage of differentiated parts of a system — the neural integration that allows for emotional regulation, relational flexibility, and coherent autobiographical narrative. Secure attachment is associated with high neural integration. Insecure attachment is associated with various forms of integration failure. Earned security, in Siegel’s model, is the result of the integrative work of making sense of one’s history — what he calls “making narrative sense” of early experience — combined with new relational experiences that provide the nervous system with the data it needs to update its predictions.

How Insecure Attachment Shows Up in Driven Women

In my work with driven, ambitious women, insecure attachment rarely looks the way it does in the textbooks. It doesn’t look like obvious neediness or dramatic abandonment panic. It looks like competence. It looks like self-sufficiency. It looks like a woman who has built an extraordinary life and still feels fundamentally alone in it.

Dismissive-avoidant attachment — the pattern most common in driven women, in my clinical experience — shows up as a deep investment in self-sufficiency. The dismissive woman doesn’t need anyone. She’s fine. She’s handled it. She’s been handling it since she was seven years old, when she learned that needing her parents created problems rather than solutions. The self-sufficiency isn’t a pose — it’s a genuine nervous system strategy. Deactivating the attachment system, turning down the volume on attachment needs, is how she survived an environment where those needs weren’t met.

Maya, the pediatric surgeon we met at the beginning of this article, is a composite of many women I’ve worked with who carry this pattern. She’s not cold. She’s not unloving. She loves her partner deeply. But the moment she feels scared — the moment she’s facing the malpractice inquiry and needs someone to hold her fear with her — something in her nervous system closes. The attachment system, trained to expect disappointment or dismissal, preemptively shuts down before it can be hurt again.

Anxious-preoccupied attachment looks different but is equally common in driven women, particularly those who grew up with inconsistent caregiving. The anxiously attached driven woman is hypervigilant in relationships — always monitoring for signs of withdrawal, always working to maintain connection, always managing the emotional temperature of the room. Her drivenness may itself be partly an attachment strategy: if I’m exceptional enough, indispensable enough, impressive enough, I won’t be left. The achievement is in service of the attachment need, not the other way around.

Disorganized attachment — the most complex pattern, associated with caregiving that was frightening or abusive — shows up in driven women as a particular kind of relational chaos. The desire for closeness and the terror of it coexist. She wants intimacy desperately and sabotages it when it gets too close. She chooses partners who are unavailable, or she becomes unavailable herself at the moment of greatest potential closeness. In relational trauma recovery, this is often the pattern that requires the most careful, titrated work.

What all three insecure patterns share is this: the internal working model predicts that closeness is unsafe, that needs will go unmet, that the self is not worthy of reliable care. And because the internal working model operates below conscious awareness — in the body, in the nervous system, in the procedural memory of how relationships work — intellectual understanding doesn’t change it. You can know your attachment style perfectly and still find yourself on the other side of the glass wall on a Sunday night.

Why Insight Alone Doesn’t Change the Pattern

This is the question I hear most often from the driven women I work with: “I understand my attachment style. I’ve read all the books. I know exactly why I do what I do. So why can’t I stop doing it?”

The answer lies in how the brain stores different types of memory. Bessel van der Kolk, MD, psychiatrist and trauma researcher, author of The Body Keeps the Score, has written extensively about the distinction between explicit memory (the narrative, conscious memory that can be accessed through language and reflection) and implicit memory (the procedural, body-based memory that operates below conscious awareness). Attachment patterns are stored primarily in implicit memory. They live in the body — in the way the chest tightens when a partner asks “what’s wrong?”, in the way the throat closes when vulnerability is required, in the way the nervous system goes offline when closeness feels too close.

Insight — the kind that comes from reading books, doing therapy that’s primarily talk-based, or intellectually understanding your patterns — operates in explicit memory. It can be profound and valuable. But it doesn’t automatically update implicit memory. The body doesn’t know you’ve read the books. The nervous system doesn’t care that you understand your dismissive attachment style. It’s still running the old program, because no one has given it a new one at the level where it actually lives.

This is why Janina Fisher, PhD, licensed psychologist and author of Healing the Fragmented Selves of Trauma Survivors, describes the challenge of trauma and attachment work as fundamentally a problem of parts. The part of you that understands your attachment style is not the same part that runs the glass wall. The understanding part is cognitive, verbal, reflective. The glass wall is run by a part that is much older, much faster, and much more convinced that closeness is dangerous. Fisher’s work on structural dissociation — the way trauma and attachment disruption fragment the self into parts that operate independently — explains why the cognitive understanding and the behavioral pattern can coexist without one changing the other.

Richard Schwartz, PhD, psychologist and developer of Internal Family Systems therapy, offers a complementary framework. In his model, described in No Bad Parts, the psyche is naturally composed of multiple parts, each with its own perspective and role. The part that maintains the glass wall — the manager that keeps Maya self-sufficient and invulnerable — is not a flaw. It’s a protector. It developed to keep the exile (the young Maya who needed her parents and found that need dangerous) from being hurt again. Attachment repair, in IFS terms, requires befriending the protector, not fighting it. It requires helping the system understand that the exile is safe now — that the current environment is not the same as the one the protector was built to navigate.

This is the work that Fixing the Foundations is designed to support: not just understanding the pattern, but actually updating it at the level where it lives — in the body, in the nervous system, in the implicit memory of how relationships work.

Both/And: Your Attachment Patterns Are Real — And They’re Not Your Destiny

Here’s what I want you to hold simultaneously, because both things are true and neither one cancels the other out.

Your attachment patterns are real. They’re not a story you’re telling yourself. They’re not a cognitive distortion. They’re encoded in your nervous system, in your implicit memory, in the procedural knowledge of how relationships work that you developed before you had language. The glass wall is real. The hypervigilance is real. The way your body closes when someone gets too close, or the way it floods with panic when someone pulls away — that’s real. It’s not a choice. It’s a survival strategy that was brilliant in the environment where it developed.

And your attachment patterns are not your destiny. This is the part that the research is unequivocal about: earned security is real. Adults who did not have secure attachment in childhood can develop it through therapeutic work, through corrective relational experiences, through the deliberate practice of letting themselves be known and finding that the world doesn’t end. Daniel Siegel’s research on earned security shows that adults who have done this work — who have made narrative sense of their history and had new relational experiences — show the same neural integration patterns as adults who were securely attached from birth. The brain doesn’t know the difference between original security and earned security. Both produce the same outcomes.

Leila is a 34-year-old product manager who came to therapy after her third significant relationship ended for what felt like the same reason: she’d been told, in different words, that she was “hard to reach.” She understood her anxious attachment style perfectly. She knew it came from her mother’s depression, from the years of inconsistent caregiving, from the way she’d learned to manage her mother’s moods rather than express her own needs. She’d done the reading. She’d done the journaling. She’d done everything she could think of to change the pattern, and she was still doing the same thing in relationships.

What Leila hadn’t done was let herself actually feel the fear — not think about it, not understand it, but feel it, in her body, in the presence of another person. The attachment work she needed wasn’t more insight. It was titrated exposure to the very experience she’d been avoiding: being seen, being scared, and finding that the other person was still there. That’s the corrective experience. That’s what updates the internal working model.

She’s doing that work now. It’s not comfortable. She told me recently that she feels worse before she feels better — that as she lets herself actually need her current partner, the old panic is louder, not quieter. That’s normal. The nervous system protests when you try to update its predictions. The exile comes forward when the protector steps back. The work is to stay with it — to let the new experience accumulate, to let the nervous system learn, slowly, that this is different.

Rupture precedes repair. That’s not a platitude. It’s a neurobiological reality. The nervous system has to encounter the feared experience — closeness, need, vulnerability — and find that it survives, in order to update its predictions. The path to earned security runs through the very territory the internal working model says is dangerous.

The Systemic Lens: Why Individualism Makes Attachment Repair Harder

There’s a cultural story about healing that is particularly seductive for driven women, and it goes like this: if you work hard enough, research thoroughly enough, and apply yourself with sufficient discipline, you can heal yourself. Alone. Through sheer force of will and intellectual rigor.

This story is not only wrong — it’s actively harmful for attachment repair. And it’s worth naming why, because the culture that produces driven women is the same culture that makes this story so compelling.

Co-regulation — the neurobiological process by which we regulate our nervous systems through contact with other regulated nervous systems — is not optional. It’s not a preference or a nice-to-have. It’s the mechanism by which attachment repair actually works. Deb Dana, LCSW, is explicit about this in her clinical writing: the nervous system learns safety through the experience of another regulated nervous system. You cannot think your way to a regulated nervous system. You cannot achieve your way to earned security. You need another person.

This is a profound challenge for the driven woman who has built her entire identity around not needing anyone. The very thing that attachment repair requires — relational vulnerability, the willingness to be regulated by another person’s presence — is the thing her internal working model has been protecting her from. The cultural ideal of the self-made, self-sufficient woman amplifies this challenge. It tells her that needing help is weakness. That healing alone is strength. That if she just works hard enough, she can do this without letting anyone in.

The research says otherwise. Daniel Siegel’s work on earned security consistently shows that the corrective experiences that produce it are relational — they happen in the context of relationships, not in isolation. Judith Herman, MD, psychiatrist and author of Trauma and Recovery, is unequivocal: recovery from trauma requires a relational witness. Not because the witness does the work for you, but because the nervous system learns safety through the experience of being witnessed, held, and not abandoned. You can’t give yourself that experience. You need another person to give it to you.

The individualism of our culture also makes it harder to access the community-level co-regulation that is, historically, how human nervous systems have always been regulated. We are not designed to regulate alone. We are designed to regulate in community — in the presence of others whose nervous systems are in ventral vagal, whose faces and voices and presence signal safety. The isolation of modern professional life — the long hours, the geographic mobility, the social thinness of digital connection — is itself a nervous system stressor that makes attachment repair harder.

This is not an argument for dependency. It’s an argument for interdependency — the recognition that needing other people is not a weakness but a biological reality, and that building the capacity for genuine co-regulation is one of the most important things a driven woman can do for her health, her relationships, and her career.

How to Heal: The Path Toward Earned Security

Attachment repair is not a linear process, and it doesn’t happen on a timeline that respects your schedule. But it does follow a recognizable arc, and understanding that arc can help you navigate it with more patience and less self-judgment.

The first step is what Judith Herman calls establishing safety — creating the conditions in which the nervous system can begin to risk new relational experiences. For many driven women, this means addressing the chronic hyperarousal or dorsal vagal shutdown that has become their baseline. It means building a nervous system regulation practice — not as a self-improvement project, but as the literal prerequisite for attachment work. You can’t update an internal working model when your nervous system is in threat response. The prefrontal cortex — the part of the brain that can integrate new experience — goes offline under threat. Safety has to come first.

The second step is finding the relational containers in which the new experiences can happen. This might be a skilled therapist. It might be a therapeutic group. It might be a relationship with a partner who is genuinely securely attached and willing to do the work with you. What matters is that the container is consistent, responsive, and safe enough for the nervous system to risk being known.

The third step is the actual work of letting the new experience land — of allowing yourself to be seen, to need, to be scared, and to find that the other person is still there. This is the titrated exposure that Peter Levine describes: small, manageable doses of the feared experience, in a context that is safe enough to survive them. Not flooding. Not avoidance. Pendulation — moving toward closeness, then back to self, then toward closeness again, each time building a little more capacity.

Donald Winnicott, MD, pediatrician and psychoanalyst, offered a concept that is quietly essential to this work: the “good enough” caregiver. Winnicott argued that the child doesn’t need a perfect caregiver — they need a good enough one, one who is present and responsive most of the time, and who repairs ruptures when they occur. The same is true of the corrective relational experiences that produce earned security. They don’t need to be perfect. They need to be good enough — consistent enough, responsive enough, and willing to repair when things go wrong. Rupture and repair is itself the mechanism of attachment learning. The repair is as important as the rupture.

For Maya, the pediatric surgeon, the work began with something small: telling her partner that she was scared about the malpractice inquiry. Not the whole story. Not a deep dive into her childhood. Just: “I’m scared about this thing at work, and I don’t know what to do.” Her partner held it well. He didn’t fix it. He didn’t minimize it. He just said, “I’m glad you told me.” And something in Maya’s nervous system — something that had been braced for decades — shifted slightly. Not dramatically. Not permanently. But enough to begin.

That’s how attachment repair works. Not in revelations, but in small moments of risk and response. Not in understanding, but in experience. Not alone, but in the presence of another person who is willing to be there.

If you’re a driven woman who recognizes herself in this article — who understands her attachment patterns intellectually and is ready to do something about them at the level where they actually live — Fixing the Foundations is the structured container I’ve built for exactly this work. It’s available self-paced at $997 or as a live cohort at $1,997. It’s built on the clinical frameworks described in this article — Judith Herman’s three-stage model, IFS parts work, somatic experiencing, and polyvagal-informed approaches — and it’s designed for the driven woman who is ready to close the gap between understanding and actually changing.

You’ve been managing this alone for a long time. You don’t have to keep doing that.

Related Reading

• Bowlby, John. Attachment and Loss, Vol. 1: Attachment. Basic Books, 1969.
• Herman, Judith Lewis. Trauma and Recovery: The Aftermath of Violence — From Domestic Abuse to Political Terror. Basic Books, 1992.
• Siegel, Daniel J. The Developing Mind: How Relationships and the Brain Interact to Shape Who We Are. Guilford Press, 1999.
• van der Kolk, Bessel. The Body Keeps the Score: Brain, Mind, and Body in the Healing of Trauma. Viking, 2014.
• Dana, Deb. The Polyvagal Theory in Therapy: Engaging the Rhythm of Regulation. W. W. Norton & Company, 2018.
• Levine, Peter A. Waking the Tiger: Healing Trauma. North Atlantic Books, 1997.

Medical Disclaimer

Medical Disclaimer