AQA A-Level Psychology Attachment Studies: Must Know, Should Know and Could Know
One of the hardest parts of revising Attachment is deciding how many studies you actually need to learn. Textbooks contain a large number of names, dates and findings, but trying to memorise every study in equal detail is rarely an effective use of revision time.
A better strategy is to organise the research into three levels:
Must Know: studies named directly in the AQA specification or central to a major theory.
Should Know: strong research evidence that can develop evaluation and support a well-constructed AO3 paragraph.
Could Know: additional studies that can add depth, originality or a counterargument when used selectively.
This guide is aligned with the updated AQA A-level Psychology specification for first A-level assessment in 2027. The specification names Lorenz, Harlow, Ainsworth, van IJzendoorn and the English and Romanian Adoptees project. It also requires Bowlby’s maternal deprivation theory, making his 44 Thieves investigation important supporting research.
Students do not need to include every study in every essay. A study earns marks only when it is accurate, relevant and clearly linked to the question.
Must Know Attachment studies
Lorenz (1935): imprinting in goslings
Lorenz investigated imprinting, an innate form of attachment seen in some bird species. He divided a clutch of greylag goose eggs into two groups. One group hatched naturally with the mother goose, while the other hatched in an incubator where Lorenz was the first moving object they encountered.
The naturally hatched goslings followed their mother. The incubator group followed Lorenz. When the two groups were placed together, the goslings continued to follow the figure on which they had originally imprinted.
Lorenz concluded that imprinting is an innate process that takes place during a brief critical period after hatching. He also observed that early imprinting could influence later courtship and sexual behaviour.
How to use it: Lorenz provides AO1 knowledge for animal studies of attachment. It can also support the idea that attachment has an innate, biological basis. However, students should be cautious when generalising from birds to human attachment, which is far more emotionally and cognitively complex.
Harlow (1959): contact comfort
Harlow separated infant rhesus monkeys from their biological mothers and gave them two artificial surrogate mothers. One was constructed from wire and supplied milk. The other was covered in soft cloth but did not necessarily provide food.
The monkeys spent considerably more time clinging to the cloth mother, including when the wire mother supplied milk. When frightened, they sought comfort from the cloth surrogate and used it as a secure base from which to explore.
Harlow concluded that contact comfort was more important than feeding in the formation of attachment. Monkeys deprived of normal maternal care later experienced serious social and reproductive difficulties, although the severity depended partly on the length of isolation.
How to use it: Harlow challenges learning theory because attachment did not form simply through the association between the caregiver and food. The study also demonstrates the long-term importance of early emotional care. Its ethical costs were extremely high, and generalisation from rhesus monkeys to humans must still be made carefully.
Ainsworth et al. (1978): the Strange Situation
Ainsworth developed the Strange Situation as a controlled observation of infant–caregiver attachment. Infants experienced a series of episodes involving play, the arrival of a stranger, separation from the caregiver and reunion.
Researchers recorded four main behaviours:
separation anxiety
stranger anxiety
reunion behaviour
willingness to explore and use the caregiver as a secure base.
Three attachment classifications were identified. Securely attached infants explored confidently, became distressed by separation and were easily comforted at reunion. Insecure-avoidant infants showed little visible distress and tended to avoid contact at reunion. Insecure-resistant infants showed intense distress but resisted comfort when the caregiver returned.
How to use it: Ainsworth is essential AO1 for the Strange Situation and attachment types. The procedure was highly standardised and has shown good inter-rater reliability. However, behaviour in the Strange Situation may reflect cultural child-rearing practices or an infant’s temperament rather than attachment security alone.
Van IJzendoorn and Kroonenberg (1988): cultural variations
Van IJzendoorn and Kroonenberg conducted a meta-analysis of 32 Strange Situation samples from eight countries, involving nearly 2,000 infants.
Secure attachment was the most common classification in every country, suggesting a degree of universality. Insecure-avoidant attachment was relatively more common in some Western European samples, whereas insecure-resistant attachment was more frequent in samples from Israel and Japan.
The most important finding was that variation within cultures was approximately one-and-a-half times greater than variation between cultures. A single national average can therefore conceal major differences between regions and social groups.
How to use it: The overall dominance of secure attachment supports Bowlby’s claim that attachment has a universal biological function. However, the variations question whether the Strange Situation measures attachment in exactly the same way in every culture. The meta-analysis also combined studies with different samples and procedures, which may reduce the validity of simple cultural comparisons.
Bowlby (1944): the 44 Thieves study
Bowlby studied 44 children who had been referred to a child guidance clinic for stealing and compared them with a control group of emotionally disturbed children who had not been referred for theft. He interviewed the children and their families, while a social worker conducted separate assessments.
Fourteen of the thieves were described as affectionless psychopaths, meaning that they showed little affection, guilt or concern for other people. Twelve of these 14 children had experienced prolonged separation from their mothers during the first two years of life.
Bowlby concluded that prolonged early separation could prevent the development of a normal attachment and contribute to later emotional difficulties and delinquency.
How to use it: The findings support Bowlby’s maternal deprivation theory. However, the research was correlational, so separation cannot be shown to have caused later difficulties. Other factors, such as family conflict or disruption, may have contributed to both the separation and the child’s later behaviour. Bowlby also carried out the interviews and developed the theory, increasing the possibility of researcher bias.
Rutter and the English and Romanian Adoptees project
The English and Romanian Adoptees project followed Romanian children adopted by British families after experiencing severe institutional deprivation. Their development was compared with British adoptees who had not experienced Romanian institutional care.
At adoption, many Romanian children showed substantial physical and cognitive delay. Children adopted before approximately six months generally made very strong recoveries. Children adopted later were more likely to show persistent difficulties, including cognitive impairment, disinhibited attachment and quasi-autistic characteristics.
Disinhibited attachment involves unusually friendly, attention-seeking behaviour towards unfamiliar adults and a lack of appropriate selectivity in attachment behaviour.
How to use it: The research demonstrates that institutionalisation can have serious long-term effects but also shows that recovery is possible. The better outcomes among children adopted very early suggest a sensitive period rather than an absolute, irreversible critical period. Because the children were not randomly assigned to deprivation, individual differences and pre-institutional experiences cannot be completely controlled. Nevertheless, studying adoptees reduced the likelihood that later outcomes were simply caused by continuing hardship.
Should Know Attachment studies
Takahashi (1990): the Strange Situation in Japan
Takahashi used the Strange Situation with 60 middle-class Japanese infants and their mothers. Secure attachment remained the most common classification, but no infants were classified as insecure-avoidant and a comparatively high proportion were classified as insecure-resistant. The procedure was ended early for some infants because their separation distress became extremely intense.
Japanese infants were rarely separated from their mothers in everyday life, so the separation episodes may have been unusually stressful. Their behaviour may therefore have reflected unfamiliarity with separation rather than insecure attachment.
How to use it: Takahashi questions the imposed etic of applying a procedure developed in the United States to another culture without considering different child-rearing practices. It is effective AO3 for the cultural validity of the Strange Situation.
Grossmann and Grossmann: attachment in Germany
Grossmann and colleagues found relatively high levels of insecure-avoidant attachment in a German sample. One explanation is that traditional German child-rearing practices placed greater value on independence and discouraged clingy behaviour.
An infant who appears avoidant in the Strange Situation may therefore be displaying culturally encouraged independence rather than an insecure relationship.
How to use it: This study further challenges the cultural validity of Ainsworth’s classifications. It shows why behaviour should be interpreted within its cultural context rather than judged automatically against an American standard.
Simonella et al. (2014): attachment patterns can change
Simonella and colleagues assessed 76 Italian infants using the Strange Situation. Secure attachment remained the largest category, but the proportion was lower than in earlier Italian research, while insecure-avoidant attachment was more common.
The researchers suggested that social changes, including longer working hours and increased use of professional childcare, might have influenced attachment behaviour.
How to use it: Simonella shows that a culture is not fixed. Attachment distributions may change across generations as working patterns, childcare and family life change. This is useful when evaluating cultural research that treats an entire country as if it has one permanent style of parenting.
Zeanah et al. (2005): attachment in Romanian institutions
Zeanah and colleagues assessed 95 Romanian children aged between 12 and 31 months who had spent most of their lives in institutional care. They were compared with 50 children who had never lived in an institution. Attachment was assessed using the Strange Situation, alongside carers’ reports of social behaviour.
Only a small minority of the institutionalised children were securely attached, compared with a clear majority of the control group. Many institutionalised children displayed disinhibited attachment towards unfamiliar adults.
How to use it: Zeanah provides strong supporting evidence for the negative effects of institutionalisation. However, it is difficult to identify which aspect of institutional care caused the difficulties because deprivation may include poor nutrition, low stimulation and frequent changes of caregiver.
Bailey et al. (2007): intergenerational attachment
Bailey and colleagues studied 99 mothers and their babies. The mothers’ attachment to their own primary attachment figures was assessed, and the babies’ attachment was measured using the Strange Situation.
Mothers who reported poor attachment relationships with their own mothers were more likely to have babies who were also classified as insecurely attached.
How to use it: The association supports Bowlby’s internal working model: early attachment may create expectations that influence later caregiving and relationships. However, the evidence is correlational. Genetic temperament or shared environmental conditions could contribute to continuity across generations.
Schaffer and Emerson (1964): multiple attachments
Schaffer and Emerson followed 60 babies from working-class families in Glasgow. Mothers were interviewed regularly about separation protest and stranger anxiety. Although many babies first formed one main attachment, most went on to form several attachments.
The strongest attachments were associated more closely with sensitive responsiveness than with the amount of time spent with the infant or responsibility for physical care.
How to use it: The stages of attachment are no longer named in the updated 2027 specification. However, the study can still be used selectively to question an overly rigid interpretation of monotropy and to show the importance of responsive interaction. Students should not spend revision time learning the discontinued stages as core content.
Could Know Attachment studies
Guiton et al. (1966): can imprinting be reversed?
Guiton found that chicks exposed to distinctive rubber gloves during an early period later directed courtship behaviour towards the gloves. However, after spending time with members of their own species, their preferences became more typical.
How to use it: Guiton qualifies Lorenz’s conclusion by suggesting that the effects of imprinting are not always completely permanent. This makes an effective counterargument in an evaluation paragraph about critical periods.
Suomi and Harlow (1972): therapist monkeys
Socially deprived monkeys were placed with younger, normally socialised monkeys. The younger animals were less threatening and encouraged interaction. Over time, the deprived monkeys developed more typical social behaviour.
How to use it: The study suggests that some effects of early deprivation can be reduced through later positive social experiences. It challenges a deterministic conclusion that early deprivation inevitably produces permanent damage, although recovery depended on the severity and duration of deprivation.
Koluchová (1972–1976): the Czech twins
Koluchová reported the case of twin boys who had experienced extreme isolation and mistreatment during early childhood. After being placed in a stable, loving environment, their intellectual, linguistic, social and emotional development improved considerably.
How to use it: The twins demonstrate that substantial recovery can occur after severe early deprivation. However, a case study of two genetically related children cannot establish how typical such recovery would be, and their exact early experiences were difficult to reconstruct.
Le Mare and Audet (2006): physical recovery after adoption
Le Mare and Audet followed Romanian children adopted into Canadian families. The adoptees showed substantial physical catch-up after leaving institutional care, although some developmental difficulties remained.
How to use it: The research supports the view that an improved environment can produce considerable recovery while also showing that recovery may be incomplete. It provides a useful comparison with Rutter’s findings.
Kagan et al. (1978): the temperament hypothesis
Kagan proposed that some behaviours used to classify attachment may arise from an infant’s biologically influenced temperament. A naturally anxious or highly reactive baby may show intense distress in the Strange Situation, whereas a calmer baby may appear less affected by separation.
How to use it: The temperament hypothesis challenges the assumption that attachment type is caused entirely by caregiver sensitivity. However, later evidence suggests that attachment and temperament are related but distinct: an infant’s behaviour results from an interaction between their characteristics and the caregiver’s response.
Fox et al. (2017): the value of early foster care
Research from the Bucharest Early Intervention Project compared children who remained in institutional care with children placed in high-quality foster care. Those moved into family care generally showed better developmental outcomes, with earlier placement associated with greater benefits.
How to use it: Fox and colleagues provide practical evidence that early intervention can improve outcomes following institutionalisation. The study also supports the idea of a sensitive period: development remains open to change, but earlier supportive care tends to produce better recovery.
How should students revise these studies?
Do not try to memorise every detail on the first attempt. Use the three levels strategically.
For each Must Know study, learn:
the aim or research question
the essential procedure
one or two key findings
the conclusion
at least one strength or limitation.
For each Should Know study, learn the finding and exactly which theory, method or conclusion it supports or challenges.
For the Could Know studies, select two or three that you can use confidently. One well-explained study is worth more than several names added without development.
An effective research paragraph should follow this sequence:
Point → Evidence → Explain → Link back to the question
For example:
One limitation of the Strange Situation is that it may be culturally biased. Takahashi found no insecure-avoidant infants in a Japanese sample but a relatively high proportion of insecure-resistant infants. Japanese infants were rarely separated from their mothers, so the procedure may have produced unusually intense distress. Therefore, the Strange Situation may measure culturally unfamiliar separation rather than attachment insecurity, reducing the validity of comparisons between cultures.
The study is not included merely to show knowledge. Every sentence explains why the evidence matters.
Final revision advice
Begin with Lorenz, Harlow, Ainsworth, van IJzendoorn, Bowlby and the English and Romanian Adoptees project. Once those studies are secure, add evidence that strengthens your evaluation. You do not need an enormous collection of names; you need a carefully selected collection that you can apply accurately.
That is the difference between simply remembering a study and using research effectively in an AQA Psychology answer.