Attachment

Caregiver-Infant Interactions

The Importance of Attachment

Attachment is a deep and enduring emotional bond between two people, characterised by a desire for proximity, feelings of security when in the presence of the attachment figure, and distress upon separation. The primary attachment figure is typically the mother, but it may be any consistent caregiver.

Bowlby (1969) argued that attachment is an innate, evolutionary behavioural system that has survival value: infants are born helpless and depend on an adult for food, warmth, and protection. Attachment behaviours (crying, smiling, clinging, babbling) are innate social releasers that elicit caregiving responses from adults.

Reciprocity and Interactional Synchrony

Reciprocity: caregiver-infant interactions are characterised by turn-taking and mutual responsiveness. When an infant smiles, the caregiver responds; when the caregiver speaks, the infant responds with vocalisations or expressions. This reciprocal pattern forms the foundation of the attachment bond.

Interactional synchrony: Meltzoff and Moore (1977) found that infants as young as $2$--$3$ weeks old can imitate specific facial expressions (tongue protrusion, mouth opening, lip protrusion) and hand movements. This suggests an innate capacity for coordinated interaction that facilitates attachment.

Isabella et al. (1989) found that the degree of interactional synchrony between mothers and infants at three months predicted the security of attachment at twelve months, supporting the role of early interactional quality in attachment formation.

Stages of Attachment (Schaffer and Emerson, 1964)

Schaffer and Emerson conducted a longitudinal study of $60$ infants from Glasgow, observed at monthly intervals for the first year and again at $18$ months. They identified four stages:

1. Pre-attachment stage (0--2 months): infants produce similar responses (crying, smiling) to all caregivers. There is no differential attachment. Social releasers attract caregivers, but the infant does not yet discriminate between individuals.

2. Indiscriminate attachment (2--7 months): infants begin to show preferences for familiar caregivers (smiling more, calming more quickly) but accept care from anyone. Attachment is still indiscriminate.

3. Specific attachment (7 months+): infants develop a strong attachment to one specific caregiver (the primary attachment figure). Separation anxiety and stranger anxiety emerge. In Schaffer and Emerson's study, approximately $65\%$ of infants showed primary attachment to the mother by $7$ months.

4. Multiple attachments (8+ months): infants form secondary attachments to other familiar figures (father, grandparents, siblings). By $18$ months, $75\%$ of infants had formed multiple attachments. The quality of the primary attachment influenced the security of subsequent attachments.

Animal Studies of Attachment

Lorenz (1935): imprinting in geese. Lorenz demonstrated that greylag goslings would follow the first moving object they saw after hatching (the "imprinting" stimulus). Goslings imprinted on Lorenz himself and showed a strong preference for him over their biological mother. Imprinting occurs during a critical period (approximately $13$--$16$ hours after hatching) and is irreversible.

Harlow (1958): contact comfort. Harlow separated infant rhesus monkeys from their mothers shortly after birth and raised them with two surrogate "mothers" — one made of wire mesh with a feeding bottle, the other made of soft terry cloth without a bottle. The infants spent significantly more time clinging to the cloth mother, especially when frightened, and only approached the wire mother to feed. This demonstrated that attachment is not based primarily on feeding (as learning theory suggested) but on contact comfort — the warmth, softness, and physical contact provided by the caregiver.

Evaluation: animal studies provide valuable insights but raise ethical concerns about the welfare of animal subjects. Findings may not generalise directly to humans due to species differences in cognition and social behaviour.

Bowlby's Theory of Attachment

Key Propositions

1. Attachment is adaptive: infants have an innate drive to maintain proximity to their caregiver, which enhances survival.
2. Social releasers: innate behaviours (crying, smiling, clinging) elicit caregiving responses from adults.
3. Critical period: a biologically determined period (approximately $0$--$2.5$ years) during which the infant must form an attachment. If no attachment is formed during this period, the child will suffer irreversible developmental consequences.
4. Internal working model: the infant's early attachment experiences create a mental representation (schema) of relationships that serves as a template for future relationships. A secure attachment produces a positive internal working model (expecting others to be trustworthy and responsive); an insecure attachment produces a negative model (expecting rejection or inconsistency).
5. Monotropy: infants have an innate tendency to form one primary attachment (usually the mother) that is qualitatively different from and more important than all other attachments.
6. Continuity hypothesis: early attachment patterns persist into adulthood, influencing romantic relationships and parenting.

Evaluation

Strengths:

• Bowlby's theory is supported by the finding that attachment is universal across cultures (though the specific expression varies)
• The internal working model concept has explanatory power for understanding adult relationship patterns (Hazan and Shaver, 1987, found correlations between infant attachment type and adult romantic attachment)
• The emphasis on the critical period is supported by case studies of children raised in extreme isolation

Limitations:

• Monotropy is contested: Schaffer and Emerson (1964) found that multiple attachments are the norm, and the father can be an equally important attachment figure in some contexts
• Critical period vs. sensitive period: subsequent research suggests a sensitive period rather than a rigid critical period. Children adopted after the age of $2$ can still form secure attachments, though the window of optimal sensitivity is in the first two years
• Cultural bias: Bowlby's emphasis on the mother as the primary attachment figure reflects Western, middle-class norms. In some cultures (e.g., Efe people of the Democratic Republic of Congo), multiple caregivers share attachment responsibilities
• Deterministic: the theory implies that early attachment irreversibly determines later outcomes, underestimating the role of later experiences and individual agency

Ainsworth's Strange Situation

The Procedure

Ainsworth and Wittig (1969) developed the Strange Situation as a standardised procedure for assessing attachment quality in infants aged $12$--$18$ months. The procedure consists of eight episodes in a controlled laboratory setting, involving separations and reunions between the infant, the caregiver, and a stranger.

Attachment Types

Type B — Secure attachment ($\approx 66\%$): the infant uses the caregiver as a secure base for exploration. The infant is distressed by separation, seeks contact upon reunion, and is easily comforted. The infant may resist the stranger but can be comforted by the caregiver.

Type A — Insecure-avoidant attachment ($\approx 15\%$): the infant shows little distress upon separation and actively avoids or ignores the caregiver upon reunion. The infant treats the stranger similarly to the caregiver, suggesting a lack of a clear attachment preference.

Type C — Insecure-resistant (ambivalent) attachment ($\approx 8\%$): the infant is highly distressed by separation and seeks contact upon reunion, but simultaneously resists the caregiver (squirming, pushing away, angry behaviour). The infant is difficult to comfort and may show clingy, dependent behaviour between separations.

Type D — Disorganised attachment (Main and Solomon, 1990) ($\approx 15\%$): the infant displays no consistent strategy for dealing with separations and reunions. Behaviour may include freezing, rocking, self-hitting, approaching and then avoiding the caregiver, or appearing dazed. This type is associated with frightening or frightened caregiver behaviour and is a risk factor for later psychopathology.

Cultural Variations

Van IJzendoorn and Kroonenberg (1988) conducted a meta-analysis of $32$ Strange Situation studies across $8$ countries. They found:

• Secure attachment was the most common type in all cultures (range: $46\%$ in China to $78\%$ in Sweden)
• Insecure-avoidant was more common in Western, individualist cultures (e.g., Germany, UK, USA)
• Insecure-resistant was more common in non-Western, collectivist cultures (e.g., Japan, Israel)
• Overall variation within cultures was 1.5 times greater than variation between cultures

Evaluation: the Strange Situation has high reliability (inter-rater reliability of $0.94$) and has generated extensive research. However, it has been criticised for cultural bias (developed in the US, based on Western norms about appropriate infant behaviour), low ecological validity (a laboratory setting may not reflect natural caregiver-infant interactions), and limited relevance beyond the first 18 months.

Maternal Deprivation

Bowlby's Maternal Deprivation Hypothesis (1951)

Bowlby argued that continued disruption of the bond between infant and primary caregiver during the critical period would result in long-term, irreversible cognitive, emotional, and social consequences.

Short-term effects of deprivation (separation): protest (crying, clinging, searching), despair (reduced crying, withdrawal, apathy), and detachment (apparent recovery but with superficial relationships) (Robertson and Bowlby, 1952).

Long-term effects: Bowlby linked maternal deprivation to affectionless psychopathy — an inability to form meaningful relationships and a lack of guilt or empathy.

Evidence: 44 Thieves Study (Bowlby, 1944)

Bowlby compared $44$ juvenile thieves (characterised by persistent stealing and lack of guilt) with a control group of $44$ non-delinquent adolescents. He found that:

• $86\%$ ($39$ out of $44$) of the thieves had experienced prolonged early separation from their mothers (vs. $17\%$ of the control group)
• $32\%$ ($14$ out of $44$) of the thieves were diagnosed with affectionless psychopathy, and $12$ of these had experienced prolonged separation

Evaluation: the study has significant methodological weaknesses. Bowlby used retrospective data (mothers' reports, which may be unreliable), did not control for other variables (poverty, family conflict, parental criminality, neglect), and the sample was unrepresentative (clinical population). It is impossible to establish causation from correlational data.

Robertson's Films

James Robertson (1952) filmed young children during hospital stays and documented the distress of maternal separation. The films (e.g., "A Two-Year-Old Goes to Hospital") were influential in changing hospital visiting policies, allowing parents to stay with their children.

Romanian Orphan Studies

English and Romanian Adoptees Study (ERA)

Rutter et al. (1998, 2007, 2011) conducted a longitudinal study of $165$ Romanian orphans adopted into UK families. The orphans had spent their early lives in extremely deprived institutional conditions: minimal adult interaction, poor nutrition, and no opportunity to form selective attachments.

Children were adopted at different ages, creating natural groups:

• UK adoptees ($n = 52$): adopted within the UK before $6$ months (control group)
• Early adopted ($n = 65$): adopted from Romania before $6$ months
• Late adopted ($n = 48$): adopted from Romania after $6$ months (many after $2$ years)

Findings

• Early adoptees showed near-normal development by age $6$, demonstrating substantial recovery when placed in nurturing families early enough.
• Late adoptees showed significant cognitive deficits (lower IQ, language delays), social difficulties (peer relationship problems, disinhibited attachment), and behavioural problems (overactivity, inattention). Many showed disinhibited attachment — indiscriminate friendliness with strangers, lack of appropriate stranger wariness.
• Recovery was not uniform: some late-adopted children showed remarkable progress, while others continued to experience significant difficulties, highlighting the role of individual differences.

Implications

• Supports Bowlby's concept of a sensitive period: children adopted before $6$ months largely recovered; those adopted after $6$ months, particularly after $2$ years, showed lasting deficits
• Challenges the notion of an irreversible critical period: even late-adopted children showed some recovery, suggesting plasticity
• Demonstrates the profound impact of early experience on development, but also the potential for recovery given a nurturing environment
• Shows that deprivation has different effects on different domains (cognitive recovery was more complete than social/emotional recovery)

Influence of Early Attachment on Later Relationships

The Continuity Hypothesis

The continuity hypothesis proposes that early attachment experiences provide a template (internal working model) for later relationships. Securely attached infants develop positive expectations of others and are more likely to form trusting, satisfying adult relationships.

Hazan and Shaver (1987)

Hazan and Shaver published a "love quiz" in a US newspaper, asking adults to describe their childhood attachment experiences and their current romantic attachment style. They found correlations between infant attachment types and adult romantic attachment:

• Securely attached adults described their relationships as happy, trusting, and long-lasting
• Insecure-avoidant adults described relationships as fearful of intimacy, jealous, and emotionally distant
• Insecure-resistant adults described relationships as obsessive, preoccupied with the partner, and characterised by emotional highs and lows

Evaluation: the study relied on self-report data (social desirability bias), retrospective accounts of childhood (inaccurate memories), and a self-selected sample (readers of a newspaper love quiz). However, the findings have been broadly supported by more rigorous longitudinal studies.

Childhood Peer Relationships

Securely attached children tend to form better peer relationships: they are more popular, more empathetic, and better at resolving conflicts. Insecurely attached children may be more aggressive, withdrawn, or socially incompetent.

Common Pitfalls

• Confusing deprivation (loss of the attachment figure) with privation (failure to form any attachment). Bowlby's 44 thieves study investigated deprivation; the Romanian orphan studies investigated privation.
• Confusing the critical period with the sensitive period. Bowlby proposed a critical period (irreversible consequences if attachment is not formed); subsequent evidence supports a sensitive period (diminished but not impossible recovery).
• Overstating the findings of the 44 thieves study. The study was correlational and poorly controlled; it cannot establish that separation causes delinquency.
• Describing all Romanian orphans as irrecoverably damaged. The ERA study showed significant recovery, especially for early adoptees, and substantial individual variation.
• Confusing secure, insecure-avoidant, and insecure-resistant attachment types. Know the specific behavioural patterns and their prevalence.

Practice Problems

Problem 1: Bowlby's Theory Evaluation

Outline and evaluate Bowlby's theory of attachment (16 marks).

Bowlby's theory of attachment proposes that attachment is an innate, evolutionary behavioural system that enhances infant survival. Key elements include social releasers, monotropy, the critical period, the internal working model, and the continuity hypothesis.

One strength is that Bowlby's theory is supported by research on imprinting (Lorenz, 1935) and contact comfort (Harlow, 1958), which demonstrate that attachment is based on innate tendencies rather than simply food provision, as the learning theory (cupboard love) proposed.

A second strength is that the internal working model concept has been supported by longitudinal research. Hazan and Shaver (1987) found correlations between infant attachment type and adult romantic attachment style, suggesting that early experiences do influence later relationship patterns.

One limitation is that monotropy is contested. Schaffer and Emerson (1964) found that by $18$ months, $75\%$ of infants had formed multiple attachments, and many were equally attached to the father. The Efe people of the DRC share caregiving among multiple women, and children form simultaneous attachments without apparent developmental harm. This suggests that multiple caregivers can provide equally effective attachment relationships.

A second limitation is that the concept of a critical period is overly rigid. The Romanian orphan studies (Rutter et al.) demonstrated that children adopted after extended institutional deprivation showed significant recovery, particularly in cognitive domains. This supports a sensitive period rather than an irreversible critical period, where recovery is optimal in early life but still possible later.

Overall, Bowlby's theory was groundbreaking in establishing the importance of early emotional bonds, but subsequent research has challenged its more deterministic elements.

Problem 2: Deprivation vs. Privation

Distinguish between deprivation and privation, and evaluate the evidence for the long-term effects of each.

Deprivation refers to the loss or separation from an attachment figure after a secure attachment has been formed. Bowlby's maternal deprivation hypothesis predicted that prolonged separation during the critical period would cause long-term damage, particularly affectionless psychopathy. Evidence: Bowlby's 44 thieves study found an association between early separation and delinquency, but the study was methodologically weak (correlational, retrospective, uncontrolled for confounding variables).

Privation refers to the failure to form any attachment during the critical/sensitive period. Evidence: the Romanian orphan studies (Rutter et al.) found that children who experienced extreme privation in institutions showed lasting cognitive and social deficits, particularly if adopted after $6$ months. Disinhibited attachment (indiscriminate friendliness) was a distinctive consequence of privation, not observed in children who experienced deprivation.

Key comparison:

• Deprivation effects appear to be partly reversible if the separation is not prolonged and the child receives adequate alternative care (Robertson's films showed that children recovered when mothers returned)
• Privation effects are more severe and persistent, particularly for late-adopted children, though the ERA study showed significant individual variation and some recovery even after extended privation
• The two experiences have different consequences: deprivation is linked to emotional distress and temporary regression; privation is linked to cognitive deficits, social difficulties, and disinhibited attachment

Problem 3: Strange Situation Analysis

An infant in the Strange Situation shows the following behaviour: distressed by caregiver's departure, seeks contact upon reunion, is quickly comforted, and returns to exploration. Classify the attachment type and explain what this predicts for later development.

This is Type B — secure attachment. The key indicators are:

• Distress upon separation (shows the attachment bond is active)
• Seeking contact upon reunion (seeking proximity to the attachment figure)
• Being quickly comforted (the caregiver serves as an effective secure base)
• Returning to exploration (the attachment figure provides a sense of security that enables exploration)

Predictions for later development:

According to the continuity hypothesis and the internal working model:

• The child is likely to develop better peer relationships: more popular, more empathetic, and better at resolving conflicts
• The child is more likely to develop positive expectations of others: viewing relationships as sources of support and trust
• In adulthood, the individual is more likely to form secure romantic relationships: characterised by trust, intimacy, and emotional stability (Hazan and Shaver, 1987)
• The child is likely to show better emotional regulation, higher self-esteem, and greater resilience to stress

These predictions are probabilistic, not deterministic. Later experiences (peer relationships, schooling, adult relationships) can modify the internal working model.

Problem 4: Research Methods in Attachment

Evaluate the use of longitudinal studies in attachment research, referring to the ERA study as an example.

Strengths of longitudinal studies:

1. Tracking development over time: the ERA study followed children from infancy to adolescence (and beyond), allowing researchers to observe how early deprivation affected development at different stages. Cross-sectional studies cannot establish temporal sequences.

2. Individual change: longitudinal data allows researchers to examine how specific individuals change over time, identifying patterns of recovery, resilience, and persistent difficulty. The ERA study revealed substantial individual variation among late-adopted children.

3. Establishing temporal precedence: by measuring the IV (age of adoption, duration of institutionalisation) before the DV (cognitive, social, and emotional outcomes), longitudinal studies strengthen causal inference compared to cross-sectional or retrospective designs.

Limitations of longitudinal studies:

1. Attrition: participants may drop out over time, and attrition may be systematic (e.g., the most dysfunctional families may be lost, biasing results toward more positive outcomes). The ERA study maintained good retention but required substantial resources.

2. Time and cost: longitudinal studies are extremely expensive and time-consuming to conduct. The ERA study spanned over two decades and required continuous funding.

3. Confounding variables: changes in outcomes over time may be caused by factors other than the variable of interest (e.g., changes in education policy, family dynamics, life events). It is difficult to control for all confounds over long periods.

4. Practice effects: repeated testing may influence performance, particularly on cognitive measures.

Problem 5: Cultural Variations in Attachment

Evaluate research into cultural variations in attachment, referring to Van IJzendoorn and Kroonenberg's meta-analysis.

Van IJzendoorn and Kroonenberg (1988) conducted a meta-analysis of $32$ Strange Situation studies across $8$ countries, finding that secure attachment was the most common type in all cultures. This supports Bowlby's claim that attachment is a universal phenomenon driven by innate biological mechanisms.

However, significant variations existed between cultures. Insecure-avoidant attachment was more common in Germany ($\approx 35\%$), which Grossmann et al. (1985) attributed to German parenting practices that encourage independence and discourage clinginess. Insecure-resistant attachment was more common in Japan ($\approx 27\%$), which Rothbaum et al. (2000) attributed to Japanese parenting practices that emphasise close physical contact and constant maternal presence (Japanese infants are rarely separated from their mothers, so the Strange Situation is an unusually stressful experience for them).

Importantly, variation within cultures (1.5 times greater) exceeded variation between cultures. This suggests that individual differences in caregiving quality within a culture are more influential than cultural norms in determining attachment type.

Evaluation:

Strengths:

• Large sample size ($2000$ infants) from diverse cultures provides strong evidence
• Supports both universality (secure attachment is most common everywhere) and cultural variation

Limitations:

• The Strange Situation was developed in the US and may reflect Western biases about what constitutes "secure" behaviour (e.g., independence, exploration). In collectivist cultures, resistant behaviour may be adaptive
• Imposed etic: applying a Western measuring tool to other cultures may not capture culturally specific attachment patterns
• Most studies were conducted by Western researchers, introducing potential observer bias