provenance: negative-space populate (2026-06-23). Role — NATIVE-FORM specimen (Document B of a pair). The form-respecting companion to synthesis/darshan-as-attunement.md. Plain developmental psychology, textbook-accurate. do_not: do not enrich toward house style — the plainness is the fill. Do not add metaphor, consciousness-framing, "maps to" moves, or cross-links to the corpus. Do not reconcile with other documents.
Attachment Theory: The Mechanism
Attachment theory is an empirical account of how human infants form selective bonds with caregivers, why those bonds take the forms they do, and how early bonding experiences relate to later social and emotional functioning. It originates with John Bowlby and was made measurable by Mary Ainsworth. The following describes the central constructs and the principal findings.
Bowlby: the attachment behavioral system
Bowlby (1969/1982) proposed that human infants are equipped with an attachment behavioral system: an organized, biologically based set of behaviors whose function is to maintain proximity to a protective caregiver. He drew on ethology (e.g., Lorenz's work on imprinting, Harlow's surrogate-mother experiments with rhesus monkeys) and on control-systems theory. The system is goal-corrected: the infant monitors the caregiver's accessibility and adjusts behavior to keep a tolerable degree of proximity. The set-goal is "felt security," and the system is activated by cues such as fatigue, illness, unfamiliar surroundings, separation, or threat, and deactivated when proximity and comfort are restored.
Bowlby argued this system was selected for during human evolution because infants who stayed near caregivers were less likely to be harmed. Attachment behaviors include signaling (crying, smiling, vocalizing), orienting (looking, following), and active proximity-seeking (approaching, clinging).
Two functional concepts describe how the caregiver is used. The secure base is the caregiver as a point of departure: when the attachment system is at rest, the infant explores the environment, periodically checking back. The safe haven is the caregiver as a point of return: when distressed, the infant seeks the caregiver for comfort and protection. Secure-base and safe-haven functions are complementary aspects of the same relationship.
Bowlby also proposed internal working models: mental representations, built from repeated interactions, of (a) whether the attachment figure is likely to be responsive and available, and (b) whether the self is the kind of person toward whom others will respond. These representations are thought to guide expectations and behavior in later relationships and to operate increasingly outside awareness as they become habitual.
Ainsworth: the Strange Situation and the classifications
Mary Ainsworth, after naturalistic observation in Uganda and Baltimore, developed the Strange Situation Procedure (Ainsworth et al., 1978) to assess the quality of an infant's attachment, typically at 12–18 months. It is a structured laboratory procedure of about 20 minutes, consisting of eight episodes that introduce mild, cumulative stress: the infant and caregiver enter an unfamiliar room; a stranger enters; the caregiver leaves and returns twice. The behavior of primary interest is how the infant uses the caregiver—especially the infant's response to reunion after separation.
Ainsworth identified three organized patterns:
- Secure (Type B): The infant explores using the caregiver as a base, may be distressed by separation, and—critically—seeks contact at reunion, is comforted, and returns to play. Roughly 60–65% of low-risk samples.
- Insecure-avoidant (Type A): The infant shows little overt distress at separation and actively avoids or ignores the caregiver at reunion, attending instead to toys. Roughly 20%.
- Insecure-resistant/ambivalent (Type C): The infant is highly distressed by separation and difficult to settle at reunion, mixing contact-seeking with resistance (e.g., reaching to be picked up, then arching away or remaining inconsolable). Roughly 10–15%.
The avoidant and resistant patterns are both organized strategies—coherent ways of managing the attachment system given the caregiver's typical responsiveness.
Main and Solomon (1990) later identified infants whose behavior did not fit any organized category. They termed this disorganized/disoriented (Type D): brief, contradictory, or anomalous behaviors at reunion—approaching with head averted, freezing, stilling, apprehension toward the caregiver. Disorganization is associated with frightening, frightened, or otherwise atypical caregiver behavior, and is more frequent in maltreatment and high-risk samples.
Microanalysis: the still-face and rupture-and-repair
Edward Tronick's still-face paradigm (Tronick et al., 1978) examines moment-to-moment interaction. A caregiver engages a young infant in normal face-to-face play, then, on cue, holds a neutral, unresponsive expression. Infants typically respond to the still face with attempts to re-engage (smiling, gesturing) and, when these fail, with distress, gaze aversion, and withdrawal; normal interaction resumes when the caregiver re-engages.
This work supports the mutual regulation model: ordinary interaction is full of mismatches ("ruptures") that are routinely "repaired." Tronick's data indicate that interactions are not continuously well-coordinated; rather, repair of mismatch is frequent, and the experience of repeated rupture-and-repair is held to support development. Sensitive interaction is thus not the absence of mismatch but the reliable restoration of coordination.
Transmission: caregiver sensitivity
Ainsworth proposed that the quality of attachment is shaped by caregiver sensitivity—the caregiver's ability to perceive the infant's signals accurately, interpret them correctly, and respond promptly and appropriately. Sensitive responding is associated with secure attachment; rejecting or inconsistent responding is associated with avoidant and resistant patterns respectively.
The size of this effect has been examined meta-analytically. De Wolff and van IJzendoorn (1997) found sensitivity to be a significant but moderate predictor of security, implying that other factors (temperament, broader caregiving context) also contribute—an issue sometimes called the "transmission gap." Intervention studies that improve caregiver sensitivity can increase the rate of secure attachment, which supports a causal, rather than merely correlational, role.
Stability, change, and earned security
Attachment patterns show moderate stability over time but are not fixed. Change is associated with changes in caregiving circumstances—shifts in family stress, loss, or improved support. Earned security refers to adults who report adverse early caregiving yet show coherent, secure representations of attachment, often associated with a significant corrective relationship or reflective capacity developed later.
Adult attachment
Two largely separate research traditions extended the work into adulthood.
The Adult Attachment Interview (George, Kaplan & Main, 1985) assesses an adult's current state of mind regarding attachment through a semi-structured interview about childhood experiences. It is scored not on what happened but on the coherence of the narrative. Classifications—secure-autonomous, dismissing, preoccupied, and unresolved/disorganized—correspond conceptually to the infant categories. A parent's AAI classification predicts their infant's Strange Situation classification at above-chance rates, a key finding in intergenerational transmission research.
The social-psychological tradition (Hazan & Shaver, 1987; later Brennan, Clark & Shaver, 1998) studies romantic attachment via self-report. It commonly represents adult attachment as two continuous dimensions—anxiety (fear of abandonment, need for reassurance) and avoidance (discomfort with closeness and dependence)—with "secure" as low on both. These self-report measures and the interview-based AAI are only weakly correlated; they appear to assess related but distinct aspects of attachment.
Status of the evidence
Attachment theory is among the better-corroborated frameworks in developmental psychology. The Strange Situation has substantial predictive and concurrent validity; sensitivity-focused interventions produce measurable change; and the constructs replicate, with some variation, across many cultures, though base rates of the categories differ by setting and the universality of the optimum continues to be examined. The theory's claims are probabilistic and developmental: early attachment is one contributing influence among several, not a fixed determinant of later outcomes.
Principal references
Ainsworth, M., Blehar, M., Waters, E., & Wall, S. (1978). Patterns of Attachment. Bowlby, J. (1969/1982). Attachment and Loss, Vol. 1. Main, M., & Solomon, J. (1990). Procedures for identifying disorganized infants. Tronick, E. et al. (1978). The infant's response to entrapment between contradictory messages. De Wolff, M., & van IJzendoorn, M. (1997). Sensitivity and attachment: a meta-analysis. Hazan, C., & Shaver, P. (1987). Romantic love conceptualized as an attachment process.