Explain how attachment develops and shapes the self, describe and evaluate Bowlby’s theory, the Strange Situation and cross-cultural research, and apply attachment theory to new situations.
Each is in the Psychology glossary, with every other term in the subject.
A close, enduring emotional bond between an infant and a caregiver, shown by seeking closeness to that person, especially when distressed, and using them as a source of comfort and security.
A characteristic pattern of attachment behaviour and expectations in close relationships, such as secure, avoidant, resistant or disorganised, in infancy or adulthood.
Bowlby’s theory that infants have an innate, evolved tendency to form an emotional bond with a caregiver, which keeps them close to protection and provides a template for later relationships.
The comfort an infant gains from soft physical contact with a caregiver, which Harlow’s research with monkeys suggested is a more important basis for attachment than feeding.
The loss of, or prolonged separation from, an attachment figure after an attachment has formed, for example through bereavement, hospitalisation or parental absence.
A pattern of indiscriminate friendliness in which a child approaches and seeks comfort from unfamiliar adults without the wariness usual for their age, often after institutional care.
An attachment type in which the infant shows contradictory, confused or frightened behaviour towards the caregiver, lacking a consistent strategy for seeking comfort.
Describe attachment and the behaviours that show an attachment has formed.
Explain Bowlby’s attachment theory, including monotropy, the secure base and internal working models.
Describe the Strange Situation and the four attachment types it identifies.
Explain the role of caregiver sensitivity and temperament in the development of attachment types.
Discuss cross-cultural research on attachment and what it shows about the universality of attachment types.
Explain how attachment contributes to the development of self, including self-worth and later relationships.
Evaluate attachment research with reference to measurement, bias, causality and responsibility.
Apply attachment theory to an unfamiliar scenario in human development.
Attachment is the strong, lasting emotional bond between an infant and the people who care for them. It matters because the first relationships a child forms appear to shape how the child comes to see themselves, whether as someone worth caring for or not, and what they expect from other people in friendships, romantic relationships and, later, as parents.
Attachment is studied in the human development context as part of the development of self. Explanations draw on all three approaches: an evolved biological system that keeps infants close to carers, cognitive representations of self and others built from early experience, and cultural norms of child care that shape what “good” caregiving looks like. Research on attachment raises sharp questions of measurement, because a bond cannot be observed directly, and of bias, because the main measure was developed with middle-class American families.
With a free account, Augment remembers where you stopped and what you have practiced.
An attachment type in which the infant explores with little reference to the caregiver, shows little distress at separation and ignores or avoids the caregiver on reunion.
An attachment type in which the infant explores little, is very distressed by separation, and on reunion both seeks contact with the caregiver and resists being comforted.
A mental representation of the self, of attachment figures and of relationships, built from early caregiving experiences, that guides expectations and behaviour in later relationships.
A caregiver’s ability to notice an infant’s signals, interpret them correctly and respond to them promptly and appropriately, proposed by Ainsworth as the main cause of secure attachment.
Bowlby’s idea that an infant has an innate tendency to form one primary attachment, usually with the mother, which is more important than any other bond.
The lack of any attachment bond, because a child never had the opportunity to form one, for example in an institution with many changing carers.
An attachment type in which the infant explores freely using the caregiver as a secure base, may be distressed by separation, and seeks and is quickly comforted by the caregiver on reunion.
A caregiver whose availability gives a child enough security to explore the environment, and to whom the child returns for comfort when threatened.
A controlled observation, developed by Ainsworth, in which an infant’s responses to two separations from and reunions with a caregiver, and to a stranger, are coded to classify the type of attachment.
The view that an infant’s inborn temperament, such as how easily they become distressed, shapes their attachment behaviour more than the sensitivity of their caregiver.
Scope: An area of study in the human development context (development of self), studied at SL and HL. Questions can ask about the role of attachment in the development of self, applied to scenarios or evaluated through any of the six concepts.
Infants are born ready to interact with people, but they do not prefer particular individuals at first. Over the first year, most infants come to direct their social behaviour towards a small number of familiar carers, and these carers become the people the infant turns to when frightened, tired or ill. This selective, emotional tie is an attachment.
Attachment cannot be seen directly, so it is inferred from behaviour. Four behaviours are used as signs that an infant has formed an attachment to a specific person:
Proximity seeking: the infant tries to stay close to the caregiver, crawling after them or reaching to be picked up.
Separation distress: the infant protests, cries or searches when the caregiver leaves.
Stranger anxiety: the infant becomes wary of unfamiliar adults, usually from about seven or eight months.
Secure-base behaviour: the infant explores the environment while the caregiver is present and returns to them from time to time to check in or be comforted.
Schaffer and Emerson (1964) followed 60 infants in Glasgow, visiting them at home every month across their first year and again at 18 months. Mothers reported how their infants reacted to everyday separations and to strangers. Most infants formed their first specific attachment in the second half of the first year, and many soon formed attachments to several people, such as fathers, grandparents and older siblings. The strongest attachment was not always to the person who fed and bathed the infant most often, but to the person who responded most to the infant’s signals and played with them.
Approximate age | Phase | What the infant does |
|---|---|---|
0–6 weeks | Asocial | Responds to people and objects in similar ways; prefers faces and voices in general. |
6 weeks–7 months | Indiscriminate attachment | Prefers people to objects, smiles at familiar and unfamiliar adults, accepts comfort from anyone. |
7–9 months | Specific attachment | Shows separation distress and stranger anxiety; one or more specific carers are preferred. |
From about 10 months | Multiple attachments | Forms further attachments to other regular carers, though one bond may remain strongest. |
The ages in the table are averages from a single Scottish sample of mainly working-class families in the 1960s, measured largely through mothers’ reports. They describe a common sequence rather than fixed stages, and infants raised in households with many regular carers may form several attachments from the start.
John Bowlby, a British psychiatrist, developed attachment theory between the 1950s and 1980s. He drew on evolutionary biology, on Konrad Lorenz’s work on imprinting in geese and on his own clinical observations of children separated from their families. His central claim was that attachment is an innate behavioural system that evolved because infants who stayed close to a protective adult were more likely to survive.
In Bowlby’s account, both sides of the relationship come prepared. Infants have social releasers, such as crying, smiling and cooing, that draw care from adults, and adults are inclined to respond to them. The attachment system is switched on by threat, such as pain, tiredness or the absence of the caregiver, and switched off by closeness and comfort. Once the infant feels safe, a complementary system, exploration, takes over.
Bowlby also proposed a critical period: if no attachment had formed by about two and a half years, it would be very difficult to form one later. Later researchers softened this into a sensitive period, a time when attachments form most easily but not the only time they can form. The difference matters for children adopted after early neglect, because a sensitive period allows for recovery.
Bowlby argued that infants form one special attachment, usually to the mother, that is qualitatively different from their other bonds. He called this monotropy. The primary attachment figure was the one the infant turned to first in distress, and the relationship with that person was thought to be the model for all later relationships.
Monotropy is the most criticised part of the theory. Schaffer and Emerson found that multiple attachments were common within months of the first one. Among the Efe, a forest-dwelling foraging people of the Democratic Republic of the Congo, Tronick, Morelli and Winn (1987) observed that infants were cared for by many people from birth, and were held, comforted and even nursed by several women, not only their mothers. Bowlby’s later writing allowed for a hierarchy of attachment figures, and most researchers now accept that infants form several attachments, often with one preferred figure.
Two further ideas link attachment to the development of self. The caregiver acts as a secure base from which the infant can explore. A child who trusts that the caregiver will be available when needed feels free to move away, play and learn; a child who is unsure stays close or explores anxiously.
Through repeated experiences of being comforted or not, the child builds an internal working model: a mental representation of what relationships are like. The model has two linked parts, a view of others (are they reliable and responsive?) and a view of the self (am I someone who deserves care?). A child whose distress is met with comfort learns that others can be trusted and that they are worth caring for. A child whose distress is ignored or rejected may learn that others are unreliable and that their own needs are a burden.
The internal working model is how attachment becomes part of the self. Bowlby described it as “working” because it can be updated by new experience, but he expected early models to be resistant to change, because they shape how a person behaves and how others respond to them. The model therefore tends to confirm itself: a child who expects rejection may withdraw, which can lead others to pay them less attention.
Bowlby’s earliest work concerned the effects of separating children from their mothers. In his study of 44 juvenile thieves seen at a London child guidance clinic, published in 1944, he classified 14 of them as “affectionless”, showing little guilt or concern for others. Twelve of these 14 had experienced prolonged separations from their mothers in early childhood, compared with only two children in a control group of clinic children who had not stolen. In a 1951 report for the World Health Organization he argued that a warm, intimate and continuous relationship with a mother, or a permanent substitute, was essential for mental health.
This maternal deprivation hypothesis had a large effect on child-care policy, such as allowing parents to stay with children in hospital. It was also used to argue that mothers should not work outside the home, an application that later research did not support. The 44 thieves study was retrospective, relied on Bowlby’s own interviews and diagnoses, and could not separate the effect of separation from the family conflict and illness that often caused it.
Before Bowlby, many psychologists explained attachment through learning: the infant comes to love the mother because she provides food, a view sometimes called “cupboard love”. Harry Harlow tested this with infant rhesus monkeys in the late 1950s. Infants were raised with two artificial “mothers”: one made of bare wire, and one covered in soft cloth. In some cases only the wire mother held a feeding bottle.
The infants fed from the wire mother when they had to, but spent most of their time clinging to the cloth mother, and ran to her when frightened by a noisy mechanical toy. Harlow concluded that contact comfort, not food, was the basis of the bond. Monkeys raised without real mothers later showed severe social problems, including difficulties mating and caring for their own young, which suggested a sensitive period for social development.
YouTubeHarlow’s research supports Bowlby’s view that attachment is an innate need rather than a learned response to food. The monkeys also suffered lasting harm that would now be judged unacceptable, which is why the research is a standard example of the ethical limits of animal studies in psychology. Generalising from rhesus monkeys to humans is also uncertain, because human infants depend on carers for much longer and their bonds develop through language and shared attention as well as touch.
Michael Rutter, in Maternal Deprivation Reassessed (1972; second edition 1981), argued that Bowlby had run together two different experiences. Deprivation is the loss of, or separation from, an attachment figure after a bond has formed. Privation is never having formed an attachment at all, as can happen in severely neglectful institutions.
Rutter argued that the most serious long-term effects, such as the inability to form close relationships, came from privation rather than from separation, and that many effects blamed on separation were due to family discord. This distinction shapes how research on children in institutions is interpreted, and it is a theme of research on the role of childhood experiences more broadly.
Bowlby described attachment as universal; Mary Ainsworth, who had worked with him in London, set out to measure how attachments differ between children. She had observed mothers and infants in their homes in Uganda in the 1950s and later in Baltimore, in the United States. To compare many infants under the same conditions, she developed a standard laboratory procedure, the Strange Situation, first reported by Ainsworth and Bell (1970).
The procedure takes place in an unfamiliar playroom and is watched through a one-way mirror. It has eight short episodes that gradually raise the infant’s stress, so that the attachment system is activated in a controlled way. Observers record how much the infant explores, how they react when the caregiver leaves, how they treat the stranger and, above all, how they behave when the caregiver returns.
Behaviour in each episode is coded for proximity seeking, contact maintaining, avoidance, resistance and search. Each infant is then placed in one of the attachment types (called attachment styles when the same patterns are studied in adults). The reunion episodes carry most weight, because a brief separation is something most infants experience in everyday life, while the way they greet the returning parent shows what they expect from that relationship.
Ainsworth identified three types. A fourth was added by Main and Solomon (1990), after researchers found that some infants did not fit any of the original categories.
Type | Exploration | Separation | Stranger | Reunion | Proposed caregiving pattern |
|---|---|---|---|---|---|
Secure (B) | Explores freely, checks back with caregiver | May be distressed, can be partly comforted by stranger | Friendly when caregiver present | Seeks contact, is comforted quickly, returns to play | Sensitive and responsive |
Insecure-avoidant (A) | Explores with little reference to caregiver | Little visible distress | Treated much like the caregiver | Ignores or turns away from caregiver | Rejecting or emotionally distant |
Insecure-resistant (C) | Explores little, stays close | Very distressed | Wary or fearful | Seeks contact but resists comfort, may hit or push away | Inconsistent: sometimes responsive, sometimes not |
Disorganised (D) | No consistent pattern | Varies | Varies | Contradictory behaviour: freezing, approaching then backing away, appearing dazed | Frightening or frightened caregiving; often linked to trauma |
Securely attached infants are not necessarily the least distressed. What defines them is that the caregiver’s return resolves their distress: they seek comfort, accept it and settle. Avoidant infants often look calm, but Spangler and Grossmann (1993) found that avoidant infants’ heart rates rose during separation much as secure infants’ did, and that their cortisol levels increased after the procedure, which suggests they have learned to suppress the display of attachment needs rather than having none.
Resistant infants show the opposite strategy. They keep the attachment system switched on, staying close and protesting loudly, which makes sense if care has been unpredictable: a child who cannot count on a response increases the signal. On reunion they both cling and push away, and they are hard to settle.
The disorganised category describes infants who have no coherent strategy at all. In the presence of the caregiver they may freeze, approach with the head turned away or appear confused. Main and Hesse proposed that this happens when the caregiver is both the source of comfort and a source of fear, so the infant is caught between approaching and escaping. Disorganised attachment is much more common in samples where children have been maltreated.
YouTubeAinsworth’s explanation was the caregiver sensitivity hypothesis. In her Baltimore home observations, mothers of secure infants noticed their infants’ signals, read them accurately and responded promptly and appropriately. Mothers of avoidant infants tended to reject bids for closeness, and mothers of resistant infants were inconsistent. On this view, the attachment type is a strategy the infant has learned for getting the best care available from a particular caregiver.
Later research supports the hypothesis, but less strongly than Ainsworth claimed. De Wolff and van IJzendoorn (1997) combined 66 studies of parental behaviour and attachment. In the 21 studies that, like Ainsworth’s, used the Strange Situation and observed sensitivity, the correlation was r = 0.24. They concluded that sensitivity is an important but not exclusive cause of security, and that other aspects of parenting, such as warmth and support, matter too.
Because sensitivity was only measured and not manipulated in these studies, a correlation cannot show that it causes security. Intervention studies come closer. Bakermans-Kranenburg, van IJzendoorn and Juffer (2003) analysed 70 intervention studies: randomised programmes increased parental sensitivity (d = 0.33) and, to a smaller extent, attachment security (d = 0.20), and interventions that changed sensitivity more also changed security more. This pattern supports a causal role for sensitivity.
The main alternative is the temperament hypothesis, associated with Jerome Kagan. Infants differ from birth in how easily they become distressed and how quickly they calm down. An easily distressed infant may cry hard at separation and be difficult to settle, and so be classified as resistant, whatever the quality of care. On this view the Strange Situation partly measures the infant’s temperament rather than the relationship.
Two findings limit the temperament hypothesis. Infants are often classified differently with each parent, which would be unlikely if the type simply reflected a fixed trait of the infant. And interventions that change only the parent’s behaviour change the infant’s attachment. Most researchers now take an interactionist view: temperament affects how easy an infant is to care for and how they express distress, while sensitivity shapes whether they feel secure.
The Strange Situation is both the main measure of infant attachment and a source of evidence about how attachment types are distributed. The extended case follows it from Ainsworth’s original American research to the cross-cultural test by van IJzendoorn and Kroonenberg (1988).
Ainsworth aimed to identify individual differences in the quality of infants’ attachments in a standard setting, and to relate them to the care the infants received at home. Van IJzendoorn and Kroonenberg aimed to find out whether the distribution of attachment types found in the United States held in other countries, and whether differences between cultures were larger than differences within them.
In Ainsworth’s research, white, middle-class mothers and their infants in Baltimore, aged about one year, took part in the eight-episode procedure. Trained observers behind a one-way mirror recorded the infants’ behaviour every 15 seconds, rating proximity seeking, contact maintaining, avoidance and resistance on seven-point scales, and the profile of scores determined the attachment type. The design is a controlled observation: the setting and sequence were standardised, but the infants’ behaviour was not manipulated beyond the planned separations.
Van IJzendoorn and Kroonenberg carried out a meta-analysis. They collected 32 Strange Situation studies from eight countries: five Western countries (West Germany, Great Britain, the Netherlands, Sweden and the United States) and three non-Western samples (Israel, Japan and China). Together the studies provided 1,990 classifications. The single Chinese sample was of Chinese families living in the United States. The researchers compared the distribution of types between countries and between samples within the same country.
Ainsworth’s own Baltimore samples classified about two-thirds of infants as secure, with more avoidant than resistant infants among the rest. Across all 32 studies, 65% of infants were secure, 21% avoidant and 14% resistant. Secure attachment was the most common type in every country, and in every sample except one German sample from Bielefeld, where avoidant infants were the largest group.
The insecure types varied more. In the Western European samples, avoidant attachment was the more common insecure type; in the Israeli and Japanese samples, resistant attachment was more common and avoidance was rare. The Japanese sample of 60 infants studied by Takahashi contained no avoidant infants at all. Variation between samples within a country was nearly 1.5 times the variation between countries. Within the United States, samples ranged widely, and the US total was very close to the global distribution.
Ainsworth concluded that infants differ systematically in how they use their caregivers for security, and that these differences reflect the care they have received. Van IJzendoorn and Kroonenberg concluded that the secure pattern is the most common across cultures, consistent with Bowlby’s claim that the attachment system is universal, while the balance of insecure types varies with child-rearing practices. Because variation within countries was larger than between them, they warned against treating any country’s results as representing a single “national” attachment style.
The Strange Situation is highly standardised, so the same procedure can be compared across many studies, and trained observers agree closely on scores: Ainsworth’s coders reached inter-rater reliability correlations of about 0.98–0.99 for behaviours such as crying and exploration. Its validity is less certain. It lasts about 20 minutes in an unfamiliar room, measures the relationship with one caregiver only, and forces continuous differences into categories. Later research suggests that attachment is better described as dimensions, such as anxiety and avoidance, than as types.
The procedure was developed with American middle-class families and assumes that a brief separation is mildly stressful and that the right response to reunion is to seek comfort and then return to play. Where these assumptions do not hold, the measure becomes an imposed etic: the same behaviour may have a different meaning in another culture, a form of cultural bias. For Japanese infants who were rarely left with anyone, the separations may have been far more stressful, making resistance more likely.
Eight countries is a narrow base for a claim of universality. Africa, South Asia and Latin America were not represented, the Chinese sample lived in the United States, and most samples were urban and middle class. Later research added some of these populations. In a study of the Dogon of Mali, True, Pisani and Oumar (2001) found 67% of 42 infants secure, none avoidant, 8% resistant and 25% disorganised, which fits a culture where infants are breastfed on demand and kept close.
The procedure deliberately creates distress. Ainsworth limited it by shortening episodes when an infant became very upset, and supporters argue that the separations are no worse than those infants experience in everyday life. Parents give consent and are present behind the mirror. Using a measure of attachment to classify a child as “insecure” also carries a responsibility: labels can lead to parents being blamed, and the classification describes one relationship at one time, not a fixed quality of the child.
Sources: 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. 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._
Cross-cultural research tests whether attachment is a universal, evolved system or a pattern shaped by the way a particular culture raises its children. Research suggests that both are partly true. The tendency to form attachments and the link between responsive care and security appear in every culture studied, while the forms that care and attachment behaviour take vary.
Japan. In traditional Japanese child rearing, infants were rarely separated from their mothers, and close mutual dependence (sometimes described with the concept of amae) was valued. High rates of resistant classifications may therefore reflect unusual distress at an unfamiliar separation. Later Japanese studies with families whose care patterns had changed found distributions closer to the global average.
Northern Germany. Grossmann and colleagues found many avoidant infants in Bielefeld, where parents encouraged early self-reliance and discouraged clinging. An infant who carries on playing at reunion may be doing what the culture expects.
Israeli kibbutzim. In some kibbutzim infants slept in communal children’s houses, cared for at night by a carer who looked after many children. Kibbutz samples showed high rates of resistant attachment, which may reflect inconsistent night-time care rather than anything about the mothers.
The Dogon of Mali. True, Pisani and Oumar found no avoidant infants, consistent with care that kept infants in close contact and fed them on demand, but a quarter of infants were disorganised, linked to mothers’ frightened or frightening behaviour.
Multiple caregiving also varies. Among the Efe, infants are passed between many adults from birth, yet they still come to prefer their mothers by the end of the first year. Findings like these suggest that the system Bowlby described is universal, but that monotropy, a single exclusive bond, is a feature of some family structures rather than a human universal.
Researchers who study attachment within its cultural setting, an emic approach, argue that the meaning of “sensitive” care differs too. In many interdependent cultures, a sensitive caregiver anticipates the infant’s needs before they are signalled, while Ainsworth’s definition rewards responding to the infant’s signals. A measure that counts only the second kind of sensitivity may underestimate the quality of care elsewhere.
Attachment is studied in human development because it is thought to shape the developing self. Three links are proposed: the internal working model gives the child a view of their own worth; the secure base supports exploration and the growth of competence; and the way caregivers respond to emotions teaches the child how to understand and manage their own feelings.
A securely attached child has repeatedly experienced distress being noticed and soothed. Bowlby argued that this is how a child comes to see themselves as lovable and capable, a foundation for self-esteem. The same experiences teach emotional regulation: at first the caregiver calms the infant, and over time the child takes over the strategies, such as seeking support or reappraising a situation. Children with avoidant histories may learn to hide negative emotions; children with resistant histories may learn to amplify them to gain attention.
Longitudinal research supports a link between early security and later social competence, but it is modest. In the Minnesota Longitudinal Study of Risk and Adaptation, which followed children from birth in low-income families, infants classified as secure were rated by teachers in preschool as more confident, more empathic and more popular with peers. Effects of this kind are usually small once family circumstances are taken into account.
If the internal working model guides later relationships, attachment patterns should show some continuity from infancy into adulthood. Hazan and Shaver (1987) applied attachment theory to romantic love. In a newspaper “love quiz”, readers chose which of three descriptions best matched their feelings in close relationships; the proportions choosing secure, avoidant and anxious descriptions were similar to the proportions of infant types, and secure adults described happier, longer relationships and warmer memories of their parents. The study was correlational and relied on a self-selected sample and recall of childhood, so it shows that adult styles resemble infant types, not that they grew out of them.
Prospective studies, which measure attachment in infancy and follow the same people, test continuity directly. They give mixed results:
Study | Sample and method | Finding |
|---|---|---|
Waters et al. (2000) | 60 white middle-class infants assessed in the Strange Situation at 12 months; 50 interviewed 20 years later with the Adult Attachment Interview, by interviewers who did not know their infant classification | 72% received the same secure or insecure classification. Of those whose mothers reported negative life events such as divorce or the death of a parent, 44% changed classification, against 22% of the others. |
Lewis, Feiring and Rosenthal (2000) | 84 white middle-class children assessed at 12 months and again with the Adult Attachment Interview at 18 years | No continuity in classification from infancy to 18. Parental divorce was related to insecure attachment at 18. |
Sources: Waters et al. (2000), _Child Development, 71(3), 684–689; Lewis, Feiring & Rosenthal (2000), Child Development, 71(3), 707–720._
Read together, the studies support the idea of a working model that is stable when life is stable and revised when it is not. Continuity depends on whether the child’s caregiving environment stays the same. A secure infant whose parents divorce in conflict may become insecure; an insecure infant who later gains a reliable carer may become secure. Bowlby called the model “working” for this reason. Early attachment therefore acts as a risk or protective factor for later development rather than fixing its course.
The strongest evidence that early relationships matter comes from children who had none. Children raised in severely depriving institutions, with many changing carers and little individual attention, often fail to form a selective attachment. Some instead show disinhibited attachment: they are indiscriminately friendly, approach strangers without wariness and may wander off with an unfamiliar adult.
In the Bucharest Early Intervention Project, Zeanah et al. (2005) assessed children aged 12–31 months living in Romanian institutions and a comparison group raised in their families. The institutionalised children showed serious disturbances of attachment on every measure, and the quality of care each child received from institution staff was related to whether they had formed an organised attachment.
Among English and Romanian adoptees, children who had spent more than six months in Romanian institutions before adoption showed more disinhibited social behaviour than children adopted earlier, and the difference persisted into young adulthood. These findings support Rutter’s view that privation, not separation as such, does the most lasting damage, and they fit a sensitive period for attachment in the first year or two.
Single-approach explanations of attachment are reductionist. Bowlby’s own theory already combined approaches: an evolved biological system, organised through cognitive working models, in a social relationship. Current research adds the role of culture and of the wider family and community.
Approach | Contribution to explaining attachment | Examples |
|---|---|---|
Biological | An evolved system that keeps infants near protection; inborn temperament; stress physiology; sensitive periods in brain development | Bowlby’s evolutionary argument; Harlow’s contact comfort; physiological stress in avoidant infants; neuroplasticity and the effects of early institutional care |
Cognitive | Internal working models: mental representations of self and others that guide expectations and are updated by experience | Continuity and change in attachment classification (Waters et al., Lewis et al.); adult attachment styles |
Sociocultural | Cultural norms of child care, family structure and who counts as a caregiver; the meaning of separation and reunion | Japanese, German, kibbutz, Dogon and Efe research; enculturation of norms about independence and closeness |
The approaches interact rather than add together. An infant’s temperament (biological) affects how easy they are to care for; how a parent responds depends on their own working model of relationships (cognitive) and on cultural ideas about crying, closeness and independence (sociocultural); and the resulting care shapes the infant’s working model and, through stress physiology, their developing brain. Explaining an attachment type through any one of these alone leaves out how they influence each other.
Attachment is a construct that must be operationalised, and the Strange Situation operationalises it through a short laboratory episode with one caregiver. Its strengths are standardisation and high inter-rater reliability. Its weaknesses are that it samples a small amount of behaviour, classifies a relationship rather than a child, depends on the infant finding the procedure mildly stressful, and imposes categories on behaviour that may vary continuously. Adult measures face similar problems: a single forced-choice item, as in Hazan and Shaver’s quiz, and an interview about childhood, the Adult Attachment Interview, may not measure the same construct, which makes continuity hard to test.
Attachment theory was developed by European and American researchers working mostly with white, middle-class, two-parent families in which the mother was the main carer. This produced a focus on the mother and on one kind of sensitivity. Feminist critics argued that the maternal deprivation hypothesis was used to keep mothers at home and gave fathers almost no role. Cross-cultural research shows that the theory holds best where family structure resembles that of its original samples.
Most evidence on the origins of attachment is correlational: sensitivity and security go together (r = 0.24 across 21 studies), but the direction is uncertain, because an easy-to-soothe infant may make a parent more confident and responsive. Shared genes, family income and stress could affect both. Randomised interventions provide the strongest evidence, and they show that changing parental behaviour changes attachment, but by a small amount (d = 0.20). The most reasonable conclusion is that sensitivity is one cause among several.
Attachment research has had positive applications, such as allowing parents to stay with children in hospital, improving care in orphanages and moving children from institutions into families. It has also been misapplied, for example to blame mothers who work or to suggest that early insecurity cannot be undone. Responsible use means communicating that attachment is influenced by many people and circumstances, and that it can change.
Because working models can be revised, attachment is a target for intervention. Programmes that coach parents to notice and respond to their infant’s signals, often using video feedback, improve sensitivity and, to a lesser degree, security. The most effective interventions in Bakermans-Kranenburg et al.’s (2003) analysis used a moderate number of sessions and a clear focus on specific parenting behaviour, which suggests that change is possible without long or intensive treatment.
In Paper 1, attachment can appear in Section B as an unseen scenario in the human development context, usually asking you to apply attachment theory to explain a child’s behaviour or development. In Section C it can be the area of study for a concept-based essay; measurement, bias, causality and responsibility all fit naturally.
Attachment is a lasting emotional bond to a caregiver, inferred from proximity seeking, separation distress, stranger anxiety and secure-base behaviour.
Bowlby’s attachment theory proposes an evolved system with a sensitive period, monotropy, a secure base and internal working models of self and others.
Harlow’s monkeys preferred a cloth mother to a feeding wire mother, supporting contact comfort over feeding as the basis of attachment.
The Strange Situation classifies infants as secure, insecure-avoidant, insecure-resistant or disorganised, mainly from behaviour at reunion.
Caregiver sensitivity is linked to security (r = 0.24), and randomised interventions that raise sensitivity modestly raise security; temperament also plays a part.
Across 32 studies in eight countries, 65% of infants were secure, but avoidance was higher in Germany and resistance higher in Israel and Japan; variation within countries was larger than between them.
The Strange Situation may be an imposed etic measure in cultures where separation is rare or independence is valued.
Attachment shapes the developing self through internal working models, but continuity into adulthood is moderate and depends on life events.
Privation in institutions can lead to disinhibited attachment, supporting a sensitive period for forming attachments.
I can describe attachment and the behaviours that show it has formed.
I can explain Bowlby’s attachment theory, including monotropy, the secure base and internal working models.
I can describe the Strange Situation and the four attachment types.
I can explain the sensitivity and temperament hypotheses and the evidence that separates them.
I can discuss cross-cultural research on attachment, including van IJzendoorn and Kroonenberg’s meta-analysis.
I can explain how attachment contributes to self-worth and later relationships, and why continuity is only moderate.
I can evaluate attachment research with reference to measurement, bias, causality and responsibility.
I can apply attachment theory to an unfamiliar scenario.