You are viewing a free preview of this lesson.
Subscribe to unlock all 10 lessons in this course and every other course on LearningBro.
Bowlby's theory makes a bold developmental prediction: that the quality of our very first attachment leaves a lasting imprint on the relationships we form for the rest of our lives. The proposed mechanism is the internal working model — a cognitive template, built from the infant's earliest experiences with its primary attachment figure, that shapes how the person comes to see themselves, what they expect of others, and how they behave in friendships, romantic partnerships and eventually their own parenting. This lesson examines the evidence for and against this continuity hypothesis, drawing on research into adult romantic relationships, the intergenerational transmission of attachment, and childhood peer relationships, before weighing the substantial AO3 debate over determinism, temperament and the durability of early attachment.
Key Definition: The internal working model (IWM) is a mental representation of the first attachment relationship, formed in infancy. It comprises a model of the self (as lovable or not), a model of others (as trustworthy and available, or not), and a model of how relationships work. Bowlby argued it acts as a schema, or template, guiding expectations and behaviour in all later relationships.
This lesson covers the AQA 7182 Paper 1 Attachment requirement: the influence of early attachment on childhood and adult relationships, including the role of the internal working model. You must be able to describe (AO1) the IWM and the continuity hypothesis, and the key supporting research — Hazan and Shaver's (1987) "Love Quiz" study of adult romantic relationships, Bailey et al.'s (2007) work on the intergenerational transmission of attachment, and Myron-Wilson and Smith's (1998) study linking attachment type to bullying. For AO3 you must evaluate the predominantly correlational evidence, the problems of retrospective self-report, the temperament alternative explanation, the concept of earned security as a challenge to determinism, and the real-world implications. The lesson builds directly on Ainsworth's attachment types and Bowlby's monotropic theory, and it is the point at which the Romanian-orphan finding of persistent disinhibited attachment is generalised into a broader theory of relational continuity.
Bowlby proposed that the first relationship acts as a prototype. From the responsiveness (or unresponsiveness) of the primary attachment figure, the infant abstracts a set of expectations about intimacy that it then carries forward. A securely attached infant, whose bids for comfort were reliably met, develops an IWM of the self as worthy of care and of others as dependable; an insecurely attached infant develops a less favourable template. The continuity hypothesis is the prediction that follows: that early attachment type predicts later social and emotional functioning, with security in infancy associated with social competence and emotional resilience later on, and insecurity associated with relational difficulty.
The proposed mechanism is worth spelling out, because it is what distinguishes a psychological theory from a mere observed correlation. Bowlby argued the IWM operates largely outside conscious awareness as a schema: it shapes what a person expects of a new relationship (will this person be available and trustworthy?), how they interpret ambiguous behaviour (is a partner's quietness a sign of rejection?), and how they themselves behave (whether they seek closeness confidently, avoid it, or cling anxiously). Because expectations tend to be self-fulfilling — an anxiously clingy partner may provoke the withdrawal they fear, while a securely confident partner elicits warmth — the early template is proposed to perpetuate itself across successive relationships. This is the engine of continuity, and it is the same mechanism invoked to explain the intergenerational transmission of attachment: a parent's IWM shapes the sensitivity of their caregiving, which shapes the infant's attachment, which becomes that infant's own IWM, and so on down the generations.
| Early Attachment Type | Internal Working Model | Predicted Later Relationship Style |
|---|---|---|
| Secure (Type B) | Self: worthy of love; Others: reliable and responsive | Trusting, emotionally intimate, confident, able to depend and be depended upon |
| Insecure-avoidant (Type A) | Self: self-reliant; Others: rejecting and distant | Discomfort with closeness, avoidance of intimacy, compulsive self-sufficiency |
| Insecure-resistant (Type C) | Self: uncertain, anxious; Others: inconsistent | Clinginess, jealousy, fear of abandonment, intense but unstable relationships |
graph TD
EARLY[First attachment in infancy<br/>quality of caregiver responsiveness] --> IWM[Internal working model:<br/>self / others / relationships]
IWM --> ROM[Adult romantic relationships]
IWM --> PAR[Own parenting behaviour]
IWM --> PEER[Childhood peer relationships]
PAR --> NEXTGEN[Child's attachment type<br/>intergenerational transmission]
The landmark study linking infant attachment to adult love is Hazan and Shaver's (1987) "Love Quiz".
| Attachment style | Approx. % of respondents | Reported adult romantic experience |
|---|---|---|
| Secure | ~56% | Relationships experienced as trusting, happy and supportive; belief in enduring love; relationships tended to last longer |
| Avoidant | ~25% | Fear of intimacy; difficulty trusting and depending on partners; scepticism about lasting love |
| Resistant/anxious | ~19% | Preoccupation with relationships; jealousy and fear of abandonment; desire for a degree of closeness that could unsettle partners |
The headline point for the exam is that this is a correlation between a retrospective self-report of childhood attachment and a self-report of adult relationships — a design that supports, but cannot prove, the continuity claim.
A further prediction of the IWM is that attachment patterns are passed down the generations: a parent's own IWM shapes the sensitivity of their caregiving, which in turn shapes their infant's attachment.
Bailey et al. (2007) assessed attachment across two generations. They studied 99 mothers and their one-year-old infants. The mothers' own attachment to their primary attachment figure was assessed (using a standard adult attachment interview), while the infants' attachment was classified using the Strange Situation. The study found that mothers who reported poor attachment to their own parents were significantly more likely to have insecurely attached infants, whereas mothers with secure attachment representations tended to have securely attached infants. This supports the idea that the IWM is transmitted through caregiving behaviour, creating an intergenerational cycle.
Converging evidence comes from longitudinal work in the tradition of the Minnesota Longitudinal Study (Sroufe et al.), which followed participants from infancy into adulthood and found continuity between an individual's early attachment and the way they later related to and cared for their own children. A useful additional reference is McCarthy (1999), who studied adult women whose infant attachment had been assessed decades earlier and found that those who had been securely attached as infants tended to have the most successful adult romantic relationships and friendships, while early insecure-resistant attachment was associated with difficulties in friendships and early insecure-avoidant attachment with difficulties in romantic intimacy.
Early attachment is also proposed to influence relationships with peers in childhood. Myron-Wilson and Smith (1998) assessed attachment type and involvement in bullying in a sample of London schoolchildren using questionnaires, and reported a clear pattern:
This fits the IWM account: a child who expects rejection may pre-emptively dominate others, while a child with an anxious, dependent model may be targeted. Supporting evidence that securely attached children tend to be more socially competent and popular with peers reinforces the link, though as ever these are correlational findings open to alternative explanation.
| Researcher | Year | Domain | Key Finding |
|---|---|---|---|
| Hazan & Shaver | 1987 | Adult romance | Recalled childhood attachment correlates with adult romantic style |
| Bailey et al. | 2007 | Parenting | Mothers' attachment representations predict their infants' attachment security |
| McCarthy | 1999 | Adult relationships | Infant attachment predicts adult romantic and friendship quality |
| Myron-Wilson & Smith | 1998 | Peer relations | Attachment type linked to bullying roles (avoidant→bully, resistant→victim) |
A further extension of the continuity hypothesis links early attachment to later mental health. The proposal is that a secure IWM acts as a protective, emotion-regulating resource, whereas insecurity is a vulnerability factor for later difficulties. Longitudinal work in the tradition of the Minnesota study (Sroufe et al.) reports that infants classified as insecurely attached were, on average, more likely to experience emotional and behavioural problems in adolescence and early adulthood, even after some other risk factors were taken into account, while early security predicted greater resilience and social competence.
The link must, however, be stated cautiously. Attachment is only one risk factor among many, and any mental-health outcome reflects a web of contributors. This naturally frames the relationship within a diathesis-stress / biopsychosocial model rather than as a simple cause.
| Contributing factor | Influence on later mental health |
|---|---|
| Early attachment | Insecure attachment raises vulnerability; secure attachment is protective |
| Genetics/temperament | Heritable predisposition to anxiety, low mood or high reactivity |
| Life events | Trauma, loss and chronic stress can trigger difficulties |
| Social support | Strong later relationships can buffer early disadvantage (cf. earned security) |
The existence of earned security is especially important here: because a corrective later relationship can revise the IWM and improve outcomes, the attachment-to-mental-health pathway is probabilistic and modifiable, not deterministic — a point with direct relevance to the treatment content of the Psychopathology topic.
Key Definition: Earned security is the development of a secure attachment representation in adulthood despite an insecure childhood attachment, typically through a corrective later relationship (a partner, close friend or therapist). It is the central evidence against a strongly deterministic reading of the continuity hypothesis.
Much of the supporting evidence is correlational, so it cannot establish that early attachment causes later relationship style. Hazan and Shaver, Bailey et al. and Myron-Wilson and Smith all report associations rather than experimental effects, and association is compatible with several causal stories. The most important alternative is the temperament hypothesis: a child's innate emotionality could shape both its Strange Situation classification in infancy and its later relational behaviour, producing a correlation that has nothing to do with an internal working model. Because attachment cannot be experimentally manipulated, this confound cannot be cleanly removed, so the continuity findings — however consistent — support rather than prove the IWM account. The implication is that the hypothesis is best stated probabilistically: early attachment is one influential predictor of later relationships, not a demonstrated cause.
The reliance on retrospective self-report seriously threatens the validity of the romantic-relationship evidence. Hazan and Shaver asked adults to recall and characterise their childhood attachment, but autobiographical memory of early relationships is reconstructive and is itself likely to be coloured by a person's current relational state and mood. A respondent presently in a happy relationship may recall their childhood more warmly, manufacturing a spurious correlation in the very direction the hypothesis predicts. This matters because it means the apparent continuity could partly be an artefact of biased recall rather than evidence of a genuine developmental thread. Prospective longitudinal designs such as the Minnesota study, which measure infant attachment directly and follow participants forward, are therefore methodologically stronger, and the fact that they too find continuity is a more secure plank of support than the Love Quiz alone.
Subscribe to continue reading
Get full access to this lesson and all 10 lessons in this course.