Explanations of Attachment: Learning Theory and Bowlby

An attachment is a close, two-way emotional bond between an infant and a caregiver, in which each individual sees the other as essential for emotional security. Psychology offers two competing explanations for how this bond forms.

Learning theory argues attachment is learned through experience. The baby learns to love whoever feeds it, because feeding brings pleasure. This is nicknamed the "cupboard love" approach — the caregiver matters because they are the source of food.

Bowlby's monotropic theory argues the opposite: attachment is innate. Babies are born biologically programmed to attach, because attachment aided survival over evolutionary history.

Explanation	Core claim	Origin of attachment
Learning theory	Attachment is learned (food-based)	Nurture / experience
Bowlby's monotropic theory	Attachment is innate (survival-based)	Nature / evolution

The whole debate turns on one question: do babies attach because of food, or because of an innate drive that has nothing to do with feeding?

Classical conditioning is learning through association. Attachment forms because the caregiver becomes associated with the pleasure of being fed.

Food produces pleasure automatically, without any learning:

• Unconditioned stimulus (UCS) = food
• Unconditioned response (UCR) = pleasure

The caregiver starts out as a neutral stimulus (NS) — the baby feels nothing special towards them. But the caregiver is present every time the baby is fed. Through repeated pairing of caregiver with food, the caregiver alone comes to produce pleasure:

• Conditioned stimulus (CS) = caregiver
• Conditioned response (CR) = pleasure

Stage	Stimulus	Response
Before	Food (UCS)	Pleasure (UCR)
Before	Caregiver (NS)	No response
During	Caregiver (NS) + Food (UCS)	Pleasure (UCR)
After	Caregiver (CS)	Pleasure (CR)

Once the caregiver reliably produces pleasure on their own, that learned association is the attachment bond, according to this explanation.

Operant conditioning is learning through the consequences of behaviour. Behaviours that are rewarded (reinforced) are repeated. Dollard and Miller (1950) applied this to attachment using the idea of drive reduction.

The chain works like this:

1. A hungry baby feels an uncomfortable internal state — hunger is a primary drive.
2. The baby cries and is fed, which reduces the discomfort. This is drive reduction, and it feels rewarding.
3. Because the caregiver supplies the food that reduces the drive, the caregiver becomes a secondary reinforcer — a source of reward in their own right.
4. The baby therefore seeks to stay close to the caregiver, and this proximity-seeking is the attachment.

The reinforcement works both ways. For the baby, being fed is positive reinforcement (gaining a reward). For the caregiver, the baby stopping crying is negative reinforcement (removing something unpleasant), which reinforces the caregiver's attention.

Positive reinforcement = behaviour is repeated because it produces a reward. Negative reinforcement = behaviour is repeated because it removes something unpleasant. Both strengthen behaviour; only punishment weakens it.

Bowlby rejected learning theory. Drawing on Lorenz's work on imprinting in animals, he argued attachment is an innate, evolved system. Babies who stayed close to a caregiver were more likely to receive food and protection, so they were more likely to survive and reproduce. Attachment behaviours were therefore naturally selected.

The theory is called monotropic because of monotropy: the idea that a baby forms one special, primary attachment, usually to the mother-figure, that is more important than all others. This bond is different in quality from a baby's other attachments, not just stronger.

Bowlby proposed two principles about how this primary bond is built:

• The law of continuity — the more constant and predictable a child's care, the better the quality of their attachment.
• The law of accumulated separation — the effects of every separation from the mother-figure add up, so even small separations carry a cost.

Monotropy does not mean a baby has only one attachment. It means one attachment is primary and qualitatively special, sitting at the top of a hierarchy of relationships.

If attachment is a two-way bond, how does the baby get the caregiver to respond? Bowlby's answer is social releasers.

Social releasers are innate infant behaviours — smiling, cooing, crying, gripping — that are "cute" and hard to ignore. They are called releasers because they release (trigger) an instinctive caregiving response in adults. Attachment is a reciprocal process: the baby's releasers draw the adult in, and the adult's care builds the bond.

Bowlby also argued attachment has a critical period. This is the window during which an attachment must form:

Concept	Bowlby's claim
Timing	Roughly the first 6 months to 2 years
What must happen	The infant must form its primary attachment within this window
If it does not	The child may struggle to form attachments later in life

Bowlby saw this as a sensitive period rather than a rigid cut-off: forming an attachment becomes much harder outside the window, but not strictly impossible. Treating it as an exact deadline overstates his claim.

The first attachment does more than keep the baby safe now. Bowlby argued it forms an internal working model (IWM) — a mental template for what relationships are like, built from the child's experience of its primary caregiver.

The internal working model shapes a person across their whole life in two key ways:

• It sets expectations of other people. A child with a loving, reliable first attachment expects relationships to be loving and reliable; a child with a poor first attachment may expect others to be unavailable or unloving.
• It provides a template for parenting. People tend to base the way they care for their own children on their own internal working model — which is why attachment patterns can be passed down through generations (the continuity hypothesis).

The internal working model links the first attachment to later relationships and to how a person will one day parent. Its influence reaches far beyond infancy — that later reach is the point examiners want you to make.

Explanations of attachment must be evaluated, not just described. Four developed points:

Learning theory has weak support. Food does not appear to be the basis of attachment. Harlow's monkeys clung to a soft cloth "mother" that provided no food rather than a wire "mother" that dispensed milk, and Schaffer and Emerson found many human babies formed their main attachment to a person who did not feed them. This strongly counters the "cupboard love" account. However, conditioning may still play a smaller role, as babies could learn to associate a carer with the comfort and responsiveness they provide, not just food.

There is support for social releasers. Brazelton et al. instructed caregivers to ignore their babies' social releasers. Using a still-face procedure, they found that babies who were previously responsive became distressed and some curled up and lay motionless. This suggests social releasers really do play an important role in the attachment process, as Bowlby claimed.

There is support for the internal working model. Bailey et al. (2007) assessed 99 mothers' attachment to their own parents and their attachment to their babies. Mothers with poor attachment to their own parents were more likely to have poorly attached infants. This supports the idea that the internal working model is passed on across generations, as Bowlby predicted.

Monotropy is socially sensitive. The claim that one primary carer is crucial can place undue pressure and blame on mothers, for example if a mother works and shares care (Burman). It may also understate the value of a wider network of carers. In addition, temperament theorists such as Kagan argue that some differences Bowlby attributed to caregiving may instead reflect a child's innate temperament.

1. Confusing classical and operant conditioning

Classical conditioning is learning by association (caregiver paired with the pleasure of food). Operant conditioning is learning by consequences (feeding reduces the hunger drive, so the caregiver becomes a secondary reinforcer). Mixing these up is one of the most common errors on this topic.

2. Saying Bowlby argued for many equal attachments

Monotropy means one primary attachment that is qualitatively special. Bowlby did not claim a baby has several equally important attachments — that misreads the whole theory. He proposed a hierarchy with one bond at the top.

3. Forgetting the internal working model links to later relationships

The internal working model is not just about the first bond. Its exam value is that it becomes a template shaping later relationships and how the person parents their own children. Answers that stop at "a model of the first attachment" miss the mark.

4. Treating the critical period as an exact cut-off

Bowlby described a sensitive period of roughly 6 months to 2 years, not a hard deadline after which attachment is impossible. Writing that a child "can never" attach after 2 years overstates the claim.

5. Describing evaluation studies without making the point

Naming Harlow, Brazelton or Bailey earns little on its own. State what the study shows about the explanation (supports social releasers, counters cupboard love, supports the IWM) to turn description into AO3.