Why You Pull Away When Someone Gets Close

Pulling away when someone gets close is the signature behavior of avoidant attachment, and it is one of the most frequently misunderstood relational patterns. People on the outside of this pattern often experience it as rejection or evidence that they are unwanted. People inside the pattern often experience it as an inexplicable need for space or a sudden loss of feeling for someone they cared about moments before. Both experiences are real. Both are explained by the same mechanism: a nervous system that learned to treat closeness as a source of threat.

The pulling away is a protection strategy. It developed for a reason. Understanding that reason is not the same as excusing the impact it has on relationships, but it is a necessary first step toward changing it.

How Avoidant Attachment Forms

Avoidant attachment, first documented empirically by developmental psychologist Mary Ainsworth in her Strange Situation research in the 1970s, develops in response to early caregiving environments where emotional expression and attachment behavior were met with rejection, dismissiveness, or withdrawal. The infant or child who reaches for a caregiver and encounters emotional distance or punishment for the reaching learns, across many iterations, that expressing attachment needs is counterproductive.

The adaptation is elegant and specific: the child's attachment system remains active, because the attachment system is biologically driven and does not simply turn off, but its behavioral expression is suppressed. The child appears less distressed than securely attached children at separation, explores more independently, and avoids the caregiver on reunion even while showing elevated cortisol levels in blood and saliva samples. The suppression is behavioral, not physiological. The nervous system is still activated. It is the expression that has been muted.

This distinction matters enormously. The avoidant person is not someone who doesn't need connection. They are someone whose system was trained to function as though they don't.

The Deactivating Strategy

Philip Shaver, a psychologist at the University of California, Davis, and Mario Mikulincer, a psychologist at Bar-Ilan University in Israel, have produced an extensive body of research on adult avoidant attachment. Their work documents the specific cognitive and emotional processes that maintain avoidant patterns, which they term deactivating strategies.

Deactivating strategies include: minimizing the importance of relationships when they become demanding, suppressing emotional responses to relational cues, redirecting attention from attachment-related thoughts toward tasks or abstractions, finding fault with partners at moments of closeness, and creating distance through behavior, conflict, or emotional withdrawal. These strategies function to keep the attachment system below its activation threshold, where it would produce the emotional vulnerability that the nervous system has learned to avoid.

Shaver and Mikulincer's experimental research shows that avoidant individuals exhibit these strategies measurably under laboratory conditions: when primed with attachment-related stimuli, they show faster reaction times to avoidant words and slower reaction times to attachment-related words, and they suppress physiological responses that securely attached individuals display openly. The processing is happening. The expression is being blocked.

Kim Bartholomew's Four-Category Model

Psychologist Kim Bartholomew at Simon Fraser University expanded Ainsworth's original three-category system into a four-category model that has been widely adopted in adult attachment research. Bartholomew proposed that adult attachment could be organized along two dimensions: the internal working model of self (positive or negative) and the internal working model of others (positive or negative). This produces four attachment categories:

Secure: positive self, positive others. Closeness is comfortable, need for connection is accepted, partners are seen as reliable.

Preoccupied (equivalent to anxious/ambivalent): negative self, positive others. Intense need for validation, hyperactivation of attachment behavior, preoccupation with relationships.

Dismissing avoidant: positive self, negative others. High self-reliance, low investment in relationships emotionally, deactivation of attachment needs.

Fearful avoidant: negative self, negative others. Desire for closeness combined with fear of closeness, producing an approach-avoidance dynamic.

Bartholomew's model is useful because it explains why people who describe pulling away when someone gets close may be doing so for very different underlying reasons. The dismissing avoidant is suppressing attachment need through self-sufficiency. The fearful avoidant is trapped between wanting connection and being frightened of it. Both pull away. The internal experience and the pathway toward change are different.

Emotionally Unavailable explores what the experience of being inside this pattern actually looks like, including how it is rationalised, how it affects the people nearby, and what it requires to move differently.

John Gottman and the Withdrawal Pattern

Psychologist and researcher John Gottman, who spent decades at the University of Washington studying couples longitudinally, identified emotional withdrawal as one of the most predictive factors of relationship dissolution. In Gottman's research, the withdrawal pattern, which he studied as stonewalling, the behavioral withdrawal from emotional interaction, was associated with elevated heart rates and physiological flooding in the person who was withdrawing.

Gottman's data supports what Ainsworth and Shaver's research suggests: the avoidant or withdrawn person is not experiencing less emotional activation. They are experiencing at least as much, and often more, and the withdrawal is the nervous system's response to becoming overwhelmed. The person who appears cold or absent during conflict is frequently the person who is most physiologically activated by it.

This reframing has clinical and relational implications. The avoidant partner who goes quiet or leaves the room during an emotional exchange is not demonstrating indifference. They are demonstrating a nervous system that became overwhelmed and defaulted to its protective strategy.

If you are trying to understand whether your pulling away is rooted in avoidant attachment or in something else, the quiz at /becoming/ provides a way to identify your specific relational patterns with more clarity.

Why Intimacy Specifically Triggers This Response

Intimacy is not simply closeness. Intimacy is closeness in which vulnerability is present: where something of the self is exposed to another person's response. For people with avoidant attachment, vulnerability is specifically the state that the deactivating system is designed to prevent. The attachment system activates most strongly in response to intimacy cues, and the deactivating response activates in proportion to the intensity of the attachment activation.

This is why many avoidant people can be comfortable and even warm in early relationship stages, when intimacy is limited, and begin withdrawing precisely as the relationship deepens. The relationship becomes more important, the vulnerability increases, and the deactivating strategy intensifies in response. From the outside, this can look like the person losing interest as the relationship became more real. From the inside, it is more accurately described as the protection system activating more strongly as the stakes increased.

The Waiting Is the Wound addresses the experience of being on the other side of this pattern: the person who reaches toward someone who is pulling away, and how to understand what that reaching is actually about.

The Dismissive vs. Fearful Distinction in Practice

People with dismissive avoidant patterns often describe the withdrawal as genuinely feeling a reduction in interest or feeling. They do not experience it as suppression, because the suppression is automatic. They are not consciously aware of wanting connection and pushing it away. They experience the end state of the deactivating process: a felt sense of not needing what they have just moved away from.

People with fearful avoidant patterns experience it quite differently: as being pulled in two directions simultaneously, wanting the closeness they are moving away from and being unable to tolerate it. The fearful avoidant experience is often more emotionally distressing and more visible to the person inside it, even though it is also more confusing and less behaviorally consistent.

Both patterns share the feature of early relational environments that made emotional expression or dependency costly. They differ in what the working model of the self contains: the dismissive avoidant's self-model tends toward inflation of self-sufficiency as a defense, while the fearful avoidant's self-model tends toward negative self-perception combined with a perception of others as both desired and dangerous.

What Helps

The research literature on changing avoidant attachment points consistently to the same conditions: a relational environment that provides consistent warmth and remains available without punishing the person for their withdrawal.

Emotionally Focused Therapy (EFT), developed by Sue Johnson, is the most extensively research-supported couples therapy for avoidant attachment and has demonstrated measurable changes in attachment organization in both individual and couples formats. EFT works by helping both partners understand the cycle that the deactivating strategy is driving, reducing blame and defensiveness, and creating space for the avoidant person to encounter vulnerability without the immediate threat-response.

Individual therapy with an attachment-focused therapist provides a relational environment in which the deactivating strategy can be explored rather than simply enacted. A therapist who maintains consistent availability and does not retaliate against distance allows the nervous system to slowly build data that contradicts the prediction that closeness is dangerous.

Psychoeducation about the mechanics of the avoidant response is useful because it allows people to observe the deactivating process rather than simply being inside it. Knowing that the withdrawal is automatic rather than chosen, and understanding what is driving it, can create enough space for a different response to become possible.

The pull away is not a character verdict. It is a strategy that once served a purpose. With a relational environment that is patient enough, the strategy can be retired gradually, as the nervous system accumulates evidence that closeness is something it can afford.