Attachment theory
- Concept · Developmental psychology
- Moderate evidence
- Published October 18, 2026
- 23 sources
- 8 min read
In short
Related in this topic
On this page
Key facts
- Field
- Developmental psychology
- Originators
- John Bowlby
- First described
- 1958–1980, United Kingdom and United States
- Also known as
- Bowlby’s attachment theory
State of the evidence
Moderate evidenceMeta-analyses support the theory’s central developmental claims, with effects that are mostly small to moderate in recent syntheses. Parental sensitivity is associated with infant security but is not its only condition[1]; parents’ attachment representations predict their infants’ attachment[2]; and security is moderately stable over the first two decades of life[3]. The transmission effect is smaller in recent studies and only partly explained by sensitivity[4], and the theory’s claim to cultural universality is contested[5].
Evidence labels summarise the research literature; they are not recommendations. How labels are assigned
Attachment theory is the account, developed by John Bowlby from the 1950s onward, of why young children form a lasting emotional bond with the people who care for them and what follows when that bond is disrupted. Its central claim is that the bond rests on an evolved behavioral system whose function is protection: the infant keeps close to a small number of familiar caregivers, and the way those caregivers respond shapes the child’s expectations of relationships[6]. Mary Ainsworth’s naturalistic observations, and the laboratory procedure she built from them, gave the theory its first systematic evidence[6].
Three neighboring terms are often confused with the theory. Attachment styles are the research classifications of individual differences that grew out of it, and have their own entry. Attachment disorders, such as reactive attachment disorder, are clinical diagnoses for children who lack attachments despite the developmental capacity to form them, not variations in security[7]. Attachment parenting is a child-rearing philosophy that borrowed the name; Duschinsky describes its appeal to Bowlby’s authority as added after its ideas were already in place[8].
Origins in separation research
Bowlby trained in medicine and in psychoanalysis in London during the 1930s and came to believe, against the Kleinian emphasis on fantasy, that actual family experiences were a basic cause of emotional disturbance[6]. His early work on separated and deprived children culminated in a 1951 report for the World Health Organization, which concluded that young children need a warm, intimate and continuous relationship with a mother or permanent mother substitute[9][6]. That work is covered in the entry on John Bowlby.
The report did not explain why the relationship should matter so much. The prevailing psychoanalytic and social learning accounts treated a child’s tie to the mother as secondary to feeding; Bowlby rejected both and turned to ethology. Konrad Lorenz’s account of imprinting in geese suggested that a social bond need not depend on food, and Robert Hinde guided Bowlby through ethological principles[6].
Harry Harlow’s experiments with rhesus monkeys supplied evidence from another species. He raised infant monkeys with two inanimate surrogate mothers, one of bare wire and one covered in cloth; for four infants the cloth surrogate gave milk, and for four the wire surrogate did. The infants spent far more time on the cloth surrogate whichever one fed them, the difference persisted over 165 days of testing, and the cloth surrogate served as a source of security when the infants were placed in a strange room[10]. Harlow concluded that contact comfort, not nursing, was the variable of overwhelming importance[10]. Bowlby drew on Harlow’s monkey research in his paper on separation anxiety[6].
Bowlby set out the theory in three papers to the British Psycho-Analytical Society between 1958 and 1960, and they met strong objections from colleagues including Anna Freud[6] . The full statement came in the trilogy Attachment and Loss: Attachment (1969), Separation (1973) and Loss (1980)[11][12][13].
The attachment behavioral system
In the first volume, Bowlby replaced Freud’s drive theory with a model borrowed from control systems. He defined attachment behavior as behavior that has proximity to an attachment figure as its predictable outcome, and whose evolutionary function is protection of the infant from danger; attachment has its own motivation and is not derived from feeding or mating[6]. It is a homeostatic control system whose set goal is proximity to the caregiver. When the child perceives a threat, such as illness, fear or separation, it seeks the caregiver, with behavior ranging from visual searching to crying and clinging; the behavior stops when conditions signal safety[14].
Two functions follow from this design. Once mobile, the infant uses the attachment figure as a secure base from which to explore and as a safe haven to return to when alarmed[6]. The secure base was Ainsworth’s contribution to the theory[6]. The two are balanced against each other: a child who perceives the caregiver as near and responsive feels secure and turns to play and exploration, and exploration declines when the attachment system is activated[14][6].
Bowlby distinguished attachment from dependency, the concept social learning theory used for the same tie. Attachment is not a sign of regression or immaturity; it performs a healthy function throughout life, including adult life[6].
Internal working models
Bowlby took the term internal working model from Kenneth Craik’s work on how organisms anticipate events: an organism that builds a model of its environment and of its own actions can predict what will happen[6]. In Separation, he proposed that the working models of the self and of attachment figures are especially salient, and that the two are complementary. A child whose attachment figure has acknowledged its need for comfort, while respecting its need to explore, is likely to model the self as valued; a child whose bids for comfort have often been rejected is likely to model the self as unworthy[6][12].
In Loss, Bowlby explained why such models become stable. Habitual patterns of thought and action become less accessible to awareness, and patterns shared between two people resist change because each person expects the other’s response[6][13]. A model also guides what a person attends to, so it tends to filter out information that would revise it; Bowlby called the motivated version of this process defensive exclusion[6].
Ainsworth’s studies in Uganda and Baltimore
Mary Ainsworth worked in Bowlby’s research unit in London before moving to Uganda in the 1950s. Near Kampala she observed 26 families with unweaned infants, visiting every two weeks for two hours over as long as nine months[6]. She described three groups: securely attached infants, who cried little and explored when the mother was present; insecurely attached infants, who cried often, even when held, and explored little; and infants not yet attached to anyone in particular. Security was associated with the mother’s sensitivity to infant signals, rated from interviews[6]. The study was published as Infancy in Uganda[15] .
From 1963 Ainsworth followed 26 families in Baltimore through the first year, with 18 home visits of about four hours each, roughly 72 hours of observation per family[6]. Mothers differed markedly in how sensitively and promptly they responded to their infants, and infants whose mothers had responded promptly to crying in the first months cried less by the end of the year[6].
The Strange Situation procedure
To see how infants balance attachment and exploration under rising stress, Ainsworth devised a laboratory procedure with Barbara Wittig, the Strange Situation[6]. The procedure had precursors in earlier studies of children in unfamiliar settings, and drew on Ainsworth’s field experience in Uganda and Baltimore[16]. It lasts about 20 minutes and runs through eight episodes, including two separations from the caregiver, the second of which leaves the infant alone[6][17].
Figure 1The Strange Situation exposes a one-year-old to a playroom, a stranger and two brief separations from the caregiver, in eight episodes that last about 20 minutes in all. Classification rests mainly on the infant’s behavior in the two reunions. After Ainsworth, Blehar, Waters and Wall (1978) and Ainsworth and Bell (1970); Ainsworth’s studies used mothers.
Text description
A numbered list of eight episodes running top to bottom, each with the people present, what happens and the duration. Episode 1, 30 seconds: an observer brings caregiver and infant into a playroom and leaves. Episode 2, 3 minutes: the infant explores while the caregiver does not take part. Episode 3, 3 minutes: a stranger enters, sits silently, talks with the caregiver and then approaches the infant, after which the caregiver leaves. Episode 4, 3 minutes or less: first separation, the infant with the stranger. Episode 5, 3 minutes or more: first reunion, in which the caregiver greets and comforts the infant and then leaves again. Episode 6, 3 minutes or less: second separation, the infant alone. Episode 7, 3 minutes or less: the stranger returns while the separation continues. Episode 8, 3 minutes: second reunion, in which the caregiver picks up the infant and the stranger leaves. Separation episodes are shaded and reunion episodes highlighted; a note states that episodes timed at 3 minutes or less are cut short if the infant becomes very distressed.
As predicted, infants explored more vigorously while their mothers were present than after the stranger entered or while the mother was away[18][6]. What interested Ainsworth more was the behavior at reunion. Most infants sought contact or interaction with the returning mother, but some were angry and resisted being settled, and others avoided her even though they had searched for her while she was gone. Home observations showed that the avoidant and resistant infants had less harmonious relationships with their mothers[6]. At Bowlby’s urging, the groups were labeled with neutral letters, A, B and C, rather than evaluative names[8]. The resulting classifications, and the disorganized category added later, are described under attachment styles.
Evidence on sensitivity and later development
The theory predicts that sensitive care produces security. A meta-analysis of 66 studies (N = 4,176) found the association in the expected direction and concluded that sensitivity is an important but not exclusive condition of security; other dimensions of parenting played an equally important role[1].
The theory also predicts that security in infancy matters for later development. Insecure attachment is associated with children’s later externalizing problems (d = 0.31 across 69 samples) and, more weakly, with internalizing symptoms (d = 0.15 across 42 samples)[19][20]. These are small group-level associations, not predictions about an individual child.
Intergenerational transmission and the transmission gap
Bowlby proposed that working models pass from parent to child through the care the parent gives[6]. The Adult Attachment Interview, developed by Mary Main and colleagues, made the proposal testable by classifying parents according to how coherently they discuss their own childhood attachments, whether or not those experiences were satisfying[6].
A 1995 meta-analysis of 18 samples (N = 854) found that parents’ interview classifications predicted their infants’ Strange Situation security with a large combined effect (d = 1.06), with 75% correspondence between secure and insecure classifications in the studies where it could be computed. Parents’ interview status also predicted their responsiveness to the infant. Yet the studies gave only partial knowledge of how representations are passed on, and sensitivity explained only part of the link, a shortfall named the transmission gap[2][4] . A synthesis of 95 samples (N = 4,819) two decades later confirmed transmission but with smaller effects (r = .31 for secure-autonomous transmission, against r = .47 earlier), smaller still in unpublished and recent studies; sensitivity narrowed the gap without closing it[4].
Criticisms from temperament, culture and stability research
Temperament
One alternative explanation holds that Strange Situation behavior reflects the infant’s temperament rather than the relationship, and in 1992 Bretherton could still describe the relation between the two as conceptually unclear[6]. A meta-analysis of 109 samples (N = 11,440) found temperament weakly associated with insecurity (d = 0.14) and modestly with resistant attachment (d = 0.30), but not significantly with avoidant or disorganized attachment[21] .
Cultural variation
An aggregation of almost 2,000 Strange Situation classifications from eight countries found avoidant classifications relatively more common in Western Europe and resistant classifications relatively more common in Israel and Japan, but variation within countries was nearly 1.5 times the variation between them[22]. Rothbaum and colleagues argued that the theory’s measures of sensitivity, competence and secure base emphasize autonomy and exploration, which reflects Western values, and that these concepts are understood differently in Japan[5]. Whether the core hypotheses are universal remains disputed.
Stability from infancy to adulthood
A meta-analysis modeling longitudinal data found attachment security moderately stable across the first 19 years of life, with a pattern best fitted by a prototype model, in which early representations keep an influence alongside later experience, rather than by a model of continual revision[3]. A larger synthesis of 127 papers estimated stability at r = .39, lower over longer intervals and in samples facing adversity, and found no significant stability over intervals longer than 15 years[23]. Security in infancy is therefore a probabilistic influence on later attachment, not a fixed trait[3][23].
Sources23
Numbered in order of first citation.
- 1.De Wolff, M. S., & van Ijzendoorn, M. H. (1997). Sensitivity and attachment: A meta-analysis on parental antecedents of infant attachment. Child Development, 68(4), 571–591. https://doi.org/10.1111/j.1467-8624.1997.tb04218.x
- 2.van IJzendoorn, M. H. (1995). Adult attachment representations, parental responsiveness, and infant attachment: A meta-analysis on the predictive validity of the Adult Attachment Interview. Psychological Bulletin, 117(3), 387–403. https://doi.org/10.1037/0033-2909.117.3.387
- 3.Fraley, R. C. (2002). Attachment stability from infancy to adulthood: Meta-analysis and dynamic modeling of developmental mechanisms. Personality and Social Psychology Review, 6(2), 123–151. https://doi.org/10.1207/S15327957PSPR0602_03
- 4.Verhage, M. L., Schuengel, C., Madigan, S., Fearon, R. M. P., Oosterman, M., Cassibba, R., Bakermans-Kranenburg, M. J., & van IJzendoorn, M. H. (2016). Narrowing the transmission gap: A synthesis of three decades of research on intergenerational transmission of attachment. Psychological Bulletin, 142(4), 337–366. https://doi.org/10.1037/bul0000038
- 5.Rothbaum, F., Weisz, J., Pott, M., Miyake, K., & Morelli, G. (2000). Attachment and culture: Security in the United States and Japan. American Psychologist, 55(10), 1093–1104. https://doi.org/10.1037/0003-066X.55.10.1093
- 6.Bretherton, I. (1992). The origins of attachment theory: John Bowlby and Mary Ainsworth. Developmental Psychology, 28(5), 759–775. https://doi.org/10.1037/0012-1649.28.5.759
- 7.Zeanah, C. H., & Gleason, M. M. (2015). Annual Research Review: Attachment disorders in early childhood – clinical presentation, causes, correlates, and treatment. Journal of Child Psychology and Psychiatry, 56(3), 207–222. https://doi.org/10.1111/jcpp.12347
- 8.Duschinsky, R. (2020). Cornerstones of attachment research. Oxford University Press.
- 9.Bowlby, J. (1951). Maternal care and mental health. World Health Organization.
- 10.Harlow, H. F. (1958). The nature of love. American Psychologist, 13(12), 673–685. https://doi.org/10.1037/h0047884
- 11.Bowlby, J. (1969). Attachment and loss: Vol. 1. Attachment. Hogarth Press.
- 12.Bowlby, J. (1973). Attachment and loss: Vol. 2. Separation: Anxiety and anger. Basic Books.
- 13.Bowlby, J. (1980). Attachment and loss: Vol. 3. Loss: Sadness and depression. Basic Books.
- 14.Fraley, R. C., & Shaver, P. R. (2000). Adult romantic attachment: Theoretical developments, emerging controversies, and unanswered questions. Review of General Psychology, 4(2), 132–154. https://doi.org/10.1037/1089-2680.4.2.132
- 15.Ainsworth, M. D. S. (1967). Infancy in Uganda: Infant care and the growth of love. Johns Hopkins Press.
- 16.Van Rosmalen, L., Van der Veer, R., & Van der Horst, F. (2015). Ainsworth's Strange Situation procedure: The origin of an instrument. Journal of the History of the Behavioral Sciences, 51(3), 261–284. https://doi.org/10.1002/jhbs.21729
- 17.Ainsworth, M. D. S., Blehar, M. C., Waters, E., & Wall, S. (1978). Patterns of attachment: A psychological study of the strange situation. Lawrence Erlbaum.
- 18.Ainsworth, M. D. S., & Bell, S. M. (1970). Attachment, exploration, and separation: Illustrated by the behavior of one-year-olds in a Strange Situation. Child Development, 41(1), 49–67. https://doi.org/10.2307/1127388
- 19.Fearon, R. P., Bakermans-Kranenburg, M. J., van Ijzendoorn, M. H., Lapsley, A. M., & Roisman, G. I. (2010). The significance of insecure attachment and disorganization in the development of children’s externalizing behavior: A meta-analytic study. Child Development, 81(2), 435–456. https://doi.org/10.1111/j.1467-8624.2009.01405.x
- 20.Groh, A. M., Roisman, G. I., van Ijzendoorn, M. H., Bakermans-Kranenburg, M. J., & Fearon, R. P. (2012). The significance of insecure and disorganized attachment for children’s internalizing symptoms: A meta-analytic study. Child Development, 83(2), 591–610. https://doi.org/10.1111/j.1467-8624.2011.01711.x
- 21.Groh, A. M., Narayan, A. J., Bakermans-Kranenburg, M. J., Roisman, G. I., Vaughn, B. E., Fearon, R. M. P., & van IJzendoorn, M. H. (2017). Attachment and temperament in the early life course: A meta-analytic review. Child Development, 88(3), 770–795. https://doi.org/10.1111/cdev.12677
- 22.van IJzendoorn, M. H., & Kroonenberg, P. M. (1988). Cross-cultural patterns of attachment: A meta-analysis of the Strange Situation. Child Development, 59(1), 147–156. https://doi.org/10.2307/1130396
- 23.Pinquart, M., Feussner, C., & Ahnert, L. (2013). Meta-analytic evidence for stability in attachments from infancy to early adulthood. Attachment & Human Development, 15(2), 189–218. https://doi.org/10.1080/14616734.2013.746257
Further reading
- Bretherton, I. (1992). The origins of attachment theory: John Bowlby and Mary Ainsworth. Developmental Psychology, 28(5), 759–775. https://doi.org/10.1037/0012-1649.28.5.759
- Duschinsky, R. (2020). Cornerstones of attachment research. Oxford University Press.Open access (CC BY-NC-ND 4.0).
- Cassidy, J., & Shaver, P. R. (2016). Handbook of attachment: Theory, research, and clinical applications (3rd ed.). Guilford Press.Edited volume; Cassidy and Shaver are the editors.
Related entries 49
- Attachment stylesConcept
- John BowlbyFigure
- Mary AinsworthFigure
- PsychoanalysisTopic
- Disorganized attachmentConcept
- Mentalization-based therapy(MBT)Therapy
- Emotionally focused therapy(EFT)Therapy
- Harry HarlowFigure
- Mary MainFigure
- Adult Attachment Interview(AAI)Method
- Attachment researchTopic
- Peter FonagyFigure
- Anxious vs avoidant attachmentComparison
- Positive vs negative punishmentComparison
- Secure vs insecure attachmentComparison
- Abandonment issuesConcept
- Adverse childhood experiencesConcept
- Anxious attachmentConcept
- Avoidant attachmentConcept
- Dismissive-avoidant attachmentConcept
- Ecological systems theoryConcept
- Emotion regulationConcept
- Intergenerational traumaConcept
- Maslow’s hierarchy of needsConcept
- MentalizationConcept
- Object permanenceConcept
- Object relations theoryConcept
- Erikson’s stages of psychosocial developmentConcept
- Secure attachmentConcept
- Self psychologyConcept
- Separation anxiety (developmental)Concept
- Stranger anxietyConcept
- TemperamentConcept
- Theory of mindConcept
- Transitional objectConcept
- Trauma bondingConcept
- TriangulationConcept
- Donald WinnicottFigure
- Margaret MahlerFigure
- Paul Gilbert (psychologist)Figure
- Susan M. JohnsonFigure
- Urie BronfenbrennerFigure
- Inner child workMethod
- Bowen family systems therapyTherapy
- Compassion-focused therapy(CFT)Therapy
- Complicated grief treatment(CGT)Therapy
- Interpersonal psychotherapy(IPT)Therapy
- Schema therapy(ST)Therapy
- Developmental psychologyTopic
Cite this entry
APA 7
Psych Dictionary. (2026, October 18). Attachment theory. In Psych Dictionary. https://www.psychdictionary.com/concepts/attachment-theory
MLA 9
“Attachment Theory.” Psych Dictionary, 18 Oct. 2026, www.psychdictionary.com/concepts/attachment-theory.