Attachment style begins forming at birth and takes its primary shape during the first 12 to 18 months of life. By the time a child is roughly two years old, their attachment pattern is typically established enough to be measured through standardized assessments. But the foundations start even earlier, with the first weeks of life setting the stage for how a child learns to relate to caregivers and, eventually, to everyone else.
The Four Stages of Attachment Formation
John Bowlby, the psychologist who developed attachment theory, described four stages that unfold in sequence during infancy. The first is pre-attachment, lasting from birth to about six weeks. During this phase, newborns don’t prefer any particular person. They respond to warmth, touch, and feeding from anyone.
Between six weeks and seven months, infants enter what Bowlby called “attachment-in-the-making.” Babies start recognizing and preferring familiar people, especially primary caregivers, but they’ll still accept comfort from others without much protest. The shift is subtle: a baby might calm down faster with a parent than a stranger, or track a familiar face more intently across the room.
The next leap happens around six to eight months, when separation anxiety appears. This is the clearest sign that a specific attachment has formed. The infant now has a strong preference for their primary caregiver and becomes visibly distressed when that person leaves. Separation anxiety peaks around 18 months. A related milestone, fear of losing love and approval, begins around 12 months and peaks at 24 months. These anxieties aren’t problems to solve. They’re developmental signals that the attachment system is working.
How Caregiving Shapes Each Style
What determines whether a child develops a secure or insecure attachment isn’t any single moment. It’s the accumulated pattern of how a caregiver responds to the infant’s signals across the entire first year. Research by Mary Ainsworth, who created the standard lab assessment for attachment (typically conducted between 12 and 25 months of age), identified clear patterns linking caregiver behavior to attachment outcomes.
Infants who developed secure attachment had caregivers who responded promptly, warmly, and consistently. When the baby cried, someone came. When the baby reached out, someone was there. The key wasn’t perfection but reliability.
Avoidant attachment formed when caregivers responded slowly, minimally, or with outright rejection. These infants learned that signaling distress didn’t reliably bring comfort, so they stopped signaling. Resistant (sometimes called anxious-ambivalent) attachment developed when caregivers were inconsistent, alternating between sensitive responses and dismissive ones. The infant couldn’t predict what they’d get, so they amplified their distress signals to try to ensure a response.
One nuanced finding stands out: how a caregiver responds during moments of distress matters more than how they engage during calm, happy interactions. Infants whose mothers were warm and attentive during playtime but relatively unresponsive when the baby was upset were actually more likely to develop avoidant attachment. Comfort during distress is the specific ingredient that builds security.
What Happens in the Brain
The reason early caregiving has such lasting effects is that it physically shapes the developing brain. The structures responsible for processing emotions, regulating fear, and managing stress are deeply influenced by the quality of care an infant receives. Sensitive, responsive caregiving helps wire strong connections between the brain’s emotional center and its regulatory regions, giving the child a built-in capacity to calm down after stress.
Poor or chaotic caregiving disrupts this wiring. Research on children raised in orphanages and on animals raised with maltreating caregivers shows weakened connectivity between emotional and regulatory brain areas, leading to heightened fear responses, difficulty managing emotions, and increased risk of depression and behavioral problems later in life. Caregivers essentially serve as an external stress-regulation system for young children, dampening the infant’s stress responses until the child’s own brain can take over. This regulatory function operates strongly through childhood but diminishes by adolescence.
The brain’s stress-regulation system appears to be “set” during infancy, calibrated to match the environment the baby actually lives in. An infant in a chaotic, unpredictable household develops a hair-trigger stress response because, in that environment, hypervigilance is adaptive. The problem is that this setting tends to persist long after the environment changes.
How Stable Is Attachment Over a Lifetime?
A 20-year longitudinal study found that 72% of people received the same secure-versus-insecure classification in early adulthood that they’d been given as infants. That’s a strong degree of continuity, but it also means nearly three in ten people changed categories.
What drove the change was life experience. Among infants whose mothers reported significant negative life events (such as loss of a parent, serious illness, or family disruption), 44% shifted attachment classifications between infancy and early adulthood. Among those without such events, only 22% changed. Stability, in other words, isn’t destiny. Attachment patterns are durable but responsive to what actually happens in a person’s life.
Environmental Factors That Shift the Odds
In the general population, roughly 60 to 75% of children develop secure attachment. Studies across different countries and age groups consistently place the secure rate in this range, with anxious-ambivalent attachment accounting for about 10 to 20% and avoidant attachment around 5 to 15%.
These numbers shift dramatically under adversity. In high-risk populations exposed to severe deprivation, maltreatment, or raised by very young mothers, the secure rate drops to as low as 0 to 30%. Disorganized attachment, the most concerning pattern (where a child shows confused or contradictory responses to the caregiver), can affect 50 to 80% of children in these groups. Income, family size, parental age, education level, major stressful events, marital conflict, and unresolved parental trauma all influence attachment outcomes, usually by affecting the caregiver’s ability to respond sensitively and consistently.
A parent carrying unresolved grief or trauma is more likely to show disrupted caregiving behaviors, not out of intention but because their own stress-regulation system is compromised. This is one of the primary mechanisms through which insecure attachment passes from one generation to the next.
Can Attachment Style Change in Adulthood?
If children reach adolescence without forming a stable attachment to a caregiver, the window for the easiest, most natural attachment formation has largely closed. But “harder” is not the same as “impossible.” The brain retains the capacity to form new connections throughout life, and therapists use the concept of “earned security” to describe adults who had insecure childhoods but developed secure attachment patterns through later relationships or therapeutic work.
The process involves building the kind of consistent, emotionally responsive relationship in adulthood that was missing in childhood, whether with a partner, a therapist, or another trusted person. Over time, these experiences can reshape the emotional and stress-regulation patterns that were set early in life. The neural mechanisms behind this shift are still being mapped, but the clinical reality is well established: people do move from insecure to secure attachment, particularly when they find relationships that offer the reliability and emotional attunement their early environment lacked.