Attachment

Note: Attachment is defined as a strong, reciprocal emotional bond between two people, such as an infant and its caregiver.

• Bodily contact - physical interactions between carer and infant help to form the attachment bond, especially after birth.
• Mimicking - infants appear to have an innate ability to imitate carers' facial expressions, which suggests it is a device to aid attachment formation.
• Caregiverese - adults interacting with infants use a different vocal language that is high-pitched, song-like, slow and repetitive. This aids communication between the infant and adult, thus strengthening the attachment bond.
• Interactional synchrony - infants move to the rhythm of their carers' spoken language to create a kind of turn-taking, two-way conversation.
• Reciprocity - interactions between carers and infants result in mutual behaviour. with both parties able to produce responses from each other.

• Bodily contact - Klaus and Kennell found that mums who had greater physical contact with their newborns appeared to form stronger and closer attachment bonds.
• Mimicking - Melzoff and Moore found that infants aged 2-3 weeks could mimic an adult's specific facial and hand gestures, supporting the idea that mimicry is an innate ability that aids the formation of attachments.
• Caregiverese - Papousek found that many cultures use this rising tone of speech when commnunicating with infants, which suggests it is an innate device to facilitate the formation of attachments.
• Interactional synchrony - Condon and Sander analysed footage of infants' movements. They found that infants co-ordinated their movements to the adults' speech.
• Reciprocity - Feldman argued that reciprocity can be seen in interactions from the age of 3 months. Meltzoff and Moore found that babies would attempt to imitate gesures from as young as 12-27 days of age.

Strengths:

• There are real-life applications. Following Klaus and Kennell's findings, hospitals began keeping mothers and babies together after birth rather than keeping them in separate rooms.

Limitations:

• Caregiverese has been seen to be used by all adults when talking to all infants. This suggests it is not a tool to form attachments, although it could be said to be a tool for aiding communication between adults and infants.
• Interactional synchrony is not found in all cultures. Le Vine et al found that Kenyan mothers had little physical contact or interactions with their infants, but they still had a high proportion of secure attachments.

Note: The stages of attachment were identified by Schaffer in 1996.

• Pre-attachment phase (birth - 3 months) - Also known as the asocial phase. From six weeks of age, infants become attracted to other humans over objects and events. This is shown by them smiling at people's faces.
• Indiscriminate attachment phase (3 - 7/8 months) - Infants begin to discriminate between familiar and unfamiliar people. They smile more at known faces. However, they will still allow strangers to handle them.
• Discriminate attachment phase (7/8 months onwards) - Also known as the specific attachment phase. Infants start developing specific attachments, staying close to familiar people and showing signs of distress when separated from them. They also avoid unfamiliar people and become distressed if strangers try to handle them.
• Multiple attachments stage (9 months onwards) - Infants begin to form emotional ties with other caregivers, such as grandparents, and non-caregivers, such as other children. Attachment to mother figure remains strongest. Their fear of strangers lessens.

Strengths:

• Supporting evidence from Schaffer and Emerson's Glasgow Babies study. They conducted a longitudinal study of 60 infants and used techniques such as observations and interviews with the mothers. They found that there was a common pattern of development in attachment formation. The method has:
  • High external validity - observations were carried out in the participants' homes, so they were more likely to act naturally than if they were in a lab setting
  • High internal validity through ise of a longitudinal study and no confounding variables caused by individual differences between participants, as they used the same 60 babies throughout.

Limitations:

• Lacks population validity. The majority of the sample was working-class, so the findings may not be applicable to other social group.
• Possibility of social desirability bias. Data collected came from interviews with the mothers, which could be questionable as they may lie due to demand characteristics.

Limitations:

• The asocial phase is difficult to study. Babies are generally immobile at this stage, so there is not much observable behaviour. This means that observations cannot be relied upon so it is difficult to draw conclusions.
• The theory is ethnocentric. There are cross-cultural differences in child-rearing, which means children could form attachments differently depending on their environment. This means it is difficult to create a theory that is applicable to all cultures.
• There is difficulty regarding the way that multiple attachments are assessed. Bowlby (1969) pointed out that infants have playmates as well as attachment figures; distress shown when a playmate leaves does not necessarily mean they have an attachment to that person. This is a problem because Schaffer's theory does not distinguish between behviour shown towards secondary attachment figures and playmates.

Traditionally, fathers have had but a minor role in the parenting of their children. In the past. mothers would raise the babies while fathers would go out to work to provide resources for the family. However, society is changing, and so is the role of fathers. Nowadays, it is more common for women to have jobs and there are far more men who care for their children while the women go out to work. There are also more single fathers raising their children. This shows that father have a much bigger role in parenting now.

Bowlby believed that infants form one key attachment, usually to the mother, but he conceded that this figure could also be the father. Many researchers suggests that fathers are seen more as a playmate, but it seems that men can develop the sensitive responsiveness needed to care for a child. Several factors affect the relationship between fathers and children:

• Degree of sensitivity - sensitivity to childrens' needs increases the security of attachments.
• Type of attachment with own parents - single-parent fathers tend to form similar attachments with their children that they had with their own parents.
• Marital intimacy - degree of intimacy with partner affects the attachment with their children.
• Supportive co-parenting - the amount of support a father gives to caring for the children affects the type of attachment they form with their children.

Geiger (1996)

• Fathers' engages in more exciting play than mothers, whose play was more affectionate
• Suggests father's role is that of a playmate, not a caregiver

Hrdy (1999)

• Fathers less able to detect low levels of infant distress than mothers
• However, Lamb (1987) found that primary caregiver fathers can quickly develop the same sensitive responsiveness as typically associated with mothers

Field (1978)

• Analysis of footage of face-to-face interactions between fathers and their infants (4 months old)
• Overall, fathers engaged more in games and held their babies less
• Primary caretaker fathers engaged in more smiling and imitative behaviours than secondary caretaker fathers. Primary caregiver fathers' behaviours comparable to primary caregiver mothers'.

Limitations:

• Inconsistent findings of research into the role of the father due to researchers being interested in different questions means psychologists cannot easily discover what the role of the father actually is
• There is no explanation as to why children with no fathers do not develop differently. MacCallum and Golombok found that children growing up in single parent or same-sex parent families do not develop differently from those growing up in two-parent heterosexual families. This suggests that a father's role is not key in a child's development.
• It is difficult to draw any firm conclusions about the exact role of the father. There are numerous influences on a child that affect their emotional development and it is difficult to control these variables.
• Research into the role of the father does not explain why fathers do not generally become primary caregivers. This could either be biological - female hormones may by linked to a more nurturing drive - or it could be due to traditional gender roles which have dictated that women be the primary caregivers.

Aim: To investigate the mechanisms of imprinting

Sample: A clutch of goose eggs

Procedure: Lorenz split the clutch in two - one he left with the mother to hatch naturally, the other he hatched using an incubator, ensuring he was the first living thing they saw. He marked them to distinguish which group they belonged to. He placed them all under a box, then lifted it up and recorded their behaviour.

Findings: The naturally hatched geese followed their mother, while the incubator-hatched ones followed Lorenz. This also happened after they were released from under the box; Lorenz's geese showed no attachment towards the mother. These bonds were irreversible. Lorenz found that imprinting would occur within 12-17 hours of birth, which he called the 'critical period'.

Conclusions: Imprinting can only occur early in life and is irreversible. This suggests that imprinting is a genetically programmed process. If an animal imprints on a person, they will carry this through their lives, leading to sexual imprinting (they will try and mate with humans in adulthood).

Strengths:

• Temporal validity - animals' behaviour doesn't change over time
• No demand characteristics
• There is support from another of Lorenz's case studies. A peacock who imprinted on a giant tortoise would only show courtship behaviour towards giant tortoises in adulthood. Supports the theory of sexual imprinting.

Limitations:

• Guiton et al found that chickens could sexually imprint on rubber gloves, but in time, learnt to prefer other chickens. Suggests that imprinting is not permanent.
• Animal studies cannot be generalised to humans.
• There are ethical issues as the geese raised by Lorenz suffered psychological harm in that they were not raised naturally.

Aim: To investigate whether food or comfort is more important in forming a bond.

Sample: 16 baby rhesus monkeys

Procedure: The monkeys were separated from the mothers and placed in a cage containing two surrogate mothers, one made of wire and one covered with soft cloth. There were two conditions: in one, the wire mother dispensed milk; in the other, the cloth mother gave milk.

Findings: The baby monkeys cuddled the cloth mother in preference to the wire mother in both conditions. They only went to the wire mother when hungry in the second condition. They sought confort from the cloth mother when frightened. In a later follow-up, Harlow's monkeys were found to be more aggressive and less sociable than other monkeys. The females had problems with child-rearing, neglecting and even attacking their young.

Conclusions: Rhesus monkeys have an innate need for contact comfort. Contact comfort is associated with a greater willingness to explore and lower levels of stress. Monkeys raised without real mothers were permanently affected.

Strengths:

• The experiment was conducted in a lab setting, which means it has high internal validity due to control of confounding variables
• There are no demand characteristics
• Real life applications. Findings led to change in zoos - babies were kept with their mothers to ensure healthy development

Limitations:

• Clear ethical issues, as the monkeys suffered lifelong psychological harm
• Lab studies lack ecological validity

Note: Learning theory is an explanation of attachment and was proposed by Dollard and Miller in 1950.

Classical conditioning:

• Caregiver begins as a neutral stimulus, producing no response from the infant.
• Food is an unconditioned stimulus, naturally giving the infant pleasure. 
• When the same person feeds them over time, the baby learns to associate them with food. 
• Once conditioning has taken place, the caregiver produces a conditioned reponse of pleasure.

Operant conditioning:

• Behaviour is either repeated or not depending on its consequences
• If a behaviour is positively reinforced, the infant will repeat it. For example, they learn that crying gets them attention, so they repeat the action whenever they want something.
• If a behaviour results in negative reinforcement, the child learns not to repeat it.

Limitations:

• Counter-evidence from Lorenz's study. His geese imprinted on him before they were fed and maintained that attachment, regardless of who fed them.
• Counter-evidence from Harlow's research. Contact comfort was seen as more important to the infant monkeys, who always preferred the company of the cloth mother even if the wire mother dispensed milk.
• Counter-evidence from Schaffer and Emerson's study. 39% of babies studied formed their first attachment to someone who did not feed them. The level of sensitive responsiveness from the caregiver was the key to forming an attachment.
• Learning theories are based on animal studies, so they are reductionist. Human behaviour is more complex and is influenced by more factors than stimulus/response models take into account.

Note: Bowlby's monotropic theory is an explanation of attachment.

• Monotropy - infants form one key attachment, typically to their mother. All other attachments are secondary.
• Babies have characteristics called social releasers which 'unlock' the innate tendency of adults to care for them. These include their appearance and their behaviour (e.g. crying, cooing).
• The ability to form an attachment is innate because they make our species more likely to survive. This innate ability means babies are given food and kept safe and warm.
• Infants must form an attachment within the critical period (birth to 2 1/2 years), otherwise the child would suffer lifelong damage.
• Internal working model - through the monotropic attachment, the infant forms a schema for relationships that will form a template for all future relationships

Strengths:

• Supported by Hazan and Shaver's Love Quiz, which showed a relationship between childhood attachment type and later romantic attachments and attitudes towards love. Suggests people do have an internal working model formed during childhood.
• Support from Lorenz' study. The geese had an innate ability and need to form an attachment within their critical period.
• Support from Harlow's study. The monkeys formed an internal working model; they suffered in adult life with mating and raising children because of their lack of a real mother.

Limitations:

• Rutter (1972) argued that all attachment figures are equally important in producing an infant's internal working model.
• Focuses only on the mother - Field found that primary fathers were just as nurturing as primary mothers, leading to secure attachments in their children.

Note: The Strange Situation was designed by Mary Ainsworth in 1970.

• Aim - To observe key attachment behaviours as a means of assessing the quality of a child's attachment to a caregiver.
• Sample - 100 middle-class American infants (aged between 9-18 months) and their mothers.
• Procedure - Mothers and infants observed in 8 stages mentioned below. Used 4 criteria (separation anxiety, stranger anxiety, reunion behaviour and safe base to explore) to classify infants as having one of 3 attachment types: secure, insecure-aviodant and insecure-resistant.
  • Stage 1 - mother and child enter the playroom
  • Stage 2 - child encouraged to explore
  • Stage 3 - stranger enters and attempts to interact
  • Stage 4 - mother leaves while stranger is present
  • Stage 5 - mother enters; stranger leaves
  • Stage 6 - mother leaves; child is alone
  • Stage 7 - stranger returns
  • Stage 8 - mother returns and interacts with child

• Findings - 70% classified as having secure attachment (Type B). 15% were insecure-resistant (Type C). 15% were insecure-aviodant (Type A). The features of each are listed below.
• Conclusion -  Most US children appeared to be securely attached. The results highlight the role of the mother's behaviour in determining the quality of the attachment.

Features of the attachment types:

• Secure (B) - High level of separation anxiety when mother leaves. Avoids stranger when alone, but friendly when mother present. Happy during reunion behaviour, when mother returns. Uses mother as a safe base to explore their environment.
• Resistant (C) - Intense levels of separation anxiety. Aviodant and fearful of stranger. During reunion, child approached mother but resists contact. Explores less than other 2 types.
• Avoidant (A) - No separation or stranger anxiety. Shows little interest when mother returns. Both mother and stranger can comfort the child equally well in the end.

Strengths:

• As the research is highly operationalised, observers have a clear view of which attachment type behaviours relate to. The means the research should have high inter-observer reliability and is also replicable, so its reliability can be checked.

Weaknesses:

• Study is low in population validity as the sample was restriced to middle-class Americans, so results may not be representative of the wider population.
• Procedure is culturally biased. Criteria are based on American values, which is Eurocentric, as observations of non-Americans will be judged by American standards. The way children are raised in different cultures may result in different attachment types; this does not mean they have worse attachments.
• The research is low in ecological validity because it was conducted in an artificial setting, meaning it lacks mundane realism.
• There are ethical issues, as the infants and possibly their mothers experienced distress.

Note: This study was conducted by Van Ijzendoorn and Kroonenberg in 1988.

Aim: To investiagte whether there are cultural variations in attachment types.

Method: Meta-analysis of 32 Strange Situation studies from 8 different countries.

Procedure: Around 2000 infant-mother relationships from 8 countries were categorised by Ainsworth's attachment types. The percentage figures of the attachment types were compared.

Findings: Secure attachments were most common in every country. Insecure avoidant was second most common in most western cultures, such as Germany and the US. Insecure resistant was second most common in Japan and Israel. China had an equal number of avoidant and resistant types.

Evaluation: Because secure attachments were most common in every country studied, it could be that attachment is an innate process shared by all children. However, there are variations. Resistant types are more common in collectivist cultures like Japan where children are not often left with people they are unfamiliar with, like childminders. Avoidant types are more common in individualistic cultures like Germany, where children are encouraged to be independent.

Strengths:

• Large sample size reduces impact of anomalous results, meaning the study has good internal validity
• Supports innate basis of attachment as it found that secure attachments were most common in every country studied

Limitations:

• Inconsistent findings. Grossman et al carried out a study in Germany and found that avoidant type was more common than resistant. Suggests meta-analysis can mask the variations within a culture
• Cultural bias. Rothbaum et al argued that the Strange Situation is rooted in 'western culture' so may not be an accurate way of measuring attachment in other cultures.
• Each country was not represented equally. Only one UK study and one Swedish study was used. Only 5 of the 32 studies were from collectivist cultures. Basing judgements of an entire country on only one or two studies can result in a biased conclusion.

• Maternal deprivation: An attachment has been formed with the main caregiver but it has been disrupted by a long-term separation. Consequences of maternal deprivation include:
  • Inability to form attachments in the future
  • Affectionless psychopathy (inability to feel remorse)
  • Deliquency (behavioural problems in adolescence)
  • Problems with cognitive development
  • Depression and dwarfism
• Privation: Not having the opportunity to form an attachment in the first place
• Separation: Not being in the presence of the primary attachment figure. Short-term separations are not significant, but longer separations can lead to deprivation (where the child loses an element of their care).
• Bowlby and Robertson (1952) looked at how short-term separation affects children's emotional health. They found that infants go through three stages in response to separation:
  • Protest - child screams and protests when caregiver tries to leave.
  • Despair - protest ends and they appear calmer. Refusal of others' attempts at comfort and often seems withdrawn and uninterested in anything.
  • Detachment - child begins to engage with others and will reject the caregiver on their return, showing signs  of anger.

Aim: To examine the link between affectionless psychopathy and maternal deprivation.

Sample: 44 male criminal teenagers accused of stealing, known as 'thieves'.

Procedure: Thieves were interviewed for signs of affectionless psychopathy, characterised as a lack of affection or guilt, and lack of empathy for their victims. Their families were also interviewed to establish whether the thieves had prolonged early spearations from their mothers. A control group of non-criminal but emotionally disturbed young people was used to see how often maternal deprivation occurred in the non-criminal children.

Findings: 14 of the 44 thieves could be described as affectionless psychopaths. 12 of these 14 experienced prolonged separations from their mothers in the first 2 years of their lives. 5 of the remaining thieves had experienced separations. Of the control group, 5 of the 44 had experienced long separations.

Conclusion: Prolonged early separation/deprivation caused affectionless psychopathy.

Strengths:

• Real-life application. Parents are now encouraged to stay with children in hospital to prevent detachment.
• Supported by Goldfarb (1955). Studied 30 orphans under 12 years old. Half had been fostered under the age of 4 months while the others remained in an orphanage. Their IQ was assessed at age 12. The fostered group averaged a score of 96 while the other group averaged 68.

Limitations:

• 44 Thieves does support Bowlby's theory but it has issues. May have been affected by researcher bias as Bowlby conducted the interviews and assessments. Relies on retrospective data which may be inaccurate. Androcentric so lacks population validity. No account for the anomalies in the control group.
• Counter evidence from Lewis (1954). Partial replication of 44 Thieves study but used a smaple of 500 young people. Found that a history of prolonged separation did not predict criminality or difficulty forming close relationships. Suggests other factors may affect the outcome of maternal deprivation.

Rutter's ERA (English and Romanian Adoptee) Study (2011)

Aim: To assess the long-term effects of early institutionalisation.

Sample: 165 Romanian orphans adopted in Britain, as well as a control group of 52 British children adopted around the same time.

Procedure: Longitudinal study. The children's physical, emotional and cognitive development was assessed at 4, 6, 11 and 15 years of age.

Findings: When the Romanian adoptees arrived in Britain, they showed signs of delayed intellectual development and most were severely undernourished. At age 11 they showed different rates of recovery related to their age of adoption. The mean IQ of those adopted under 6 months of age was 102, compared with 86 for those adopted between 6 months and 2 years and 77 for those adopted after 2 years. These differences remained at the age of 16.

Conclusion: There appeared to be a difference in attachment outcomes related to whether or not adoption had took place before or after 6 months of age. Those adopted after 6 months showed signs of disinhibited attachment, which is characterised by attention seeking and clinginess directed towards any adult, both familiar and unfamiliar.

The Bucharest Early Intervention Project - Zeanah et al (2005)

Aim: To assess attachment types in children in long-term institutional care.

Sample: 95 children aged 12-31 months who had spent most of their lives (avg. 90%) in institutional care. Control group of 50 children who had never lived in an institution.

Procedure: Longitudinal study. Attachment type meadured using Strange Situation. Carers asked about unusual social behaviour, e.g. clingy behaviours directed towards any adult (i.e. disinhibited attachment).

Findings: 74% of control group was securely attached compared with only 19% of the institutional group. 65% of them had a disorganised attachment type and the description of disinhibited attachment applied to 44% of them, as opposed to less than 20% of the control group.

Conclusion: Institutional care has an impact on how a child forms attachments in later life.

Strengths:

• Real-life application. Studying Romanian orphans has led to improvements in the way children are cared for in institutions. For example, they now ensure a smaller number of people play a central role for the child to avoid disinhibited attachment.
• Less extraneous variables than other orphan studies. There were no confounding variables such as loss or trauma that the children were also dealing with.

Limitations:

• The Romanian orphanages were not typical. The conditions were so bad that the results cannot be applied to understanding the effects of better quality institutional care. This means the studies lack generalisability.
• Long-term effects are unclear. The children adopted later may have eventually caught up with those adopted earlier in adulthood, and those adopted earlier may later suffer with emotional problems. It is too soon to say if these results would be lasting.
• Ethical issues. The orphans in Zeanah et al's study were randomly assigned to institutional care or fostering, which may be methodologically better but it raises ethical issues.

Key terms:

Childhood relationships: Affiliations with other people in childhood, including friends and classmates, as well as with adults such as teachers.

Adult relationships: Those relationships the child goes on to have later in life as an adult. These include friendships and working relationships but most critically relationships with romantic partners and the person's own children.

Internal working model: Designed by Bowlby (1969). The mental representations we all carry with us of our attachment to our primary caregiver. They are important in affecting our future relationships because they carry our perception of what relationships look like, thus acting as a template for our future relationships. For example, type A relationship behaviour includes not being too emotionally close or involved, ad type C behaviour includes being controlling and argumentative.

Relationships in later childhood

• Kerns (1994) - securely attached infants tend to form the best quality childhood friendships whereas insecurely attached infants have friendship difficulties.
• Wilson and Smith (1998) - assessed attachment type and bullying involvement in 196 children aged 9-11 in London. Found that secure children were unlikely to be involved in bullying, insecure-avoidant children were most likely to be victims and insecure-resistant children were most likely to be bullies.

Romantic relationships in adulthood

• McCarthy (1999) - 40 women who had been assessed for attachment type in infancy were studied in adulthood. Those that were securely attached had the best adult relationships. Insecure-resistant ones had problems maintaining relationships, and insecure-avoidant ones struggled with intimacy in romantic relationships.
• Hazan and Shaver's Love Quiz (1987) - analysed replies to a questionnaire printed in a newspaper. Assessed attachmen type and love experiences and attitudes, Securely attached people were most likely to have good, long-lasting relationships. Insecure-avoidant people tended to reveal jealousy and fear of intimacy.

Strengths:

• Some studies support the internal working model (Hazan and Shaver, McCarthy)
  • However, not all do - Zimmerman (2000) found little relationship between infant attachment type and adolescent attachment to parents

Limitations:

• Many studies are self-report, which created internal validity problems. Social desirability bias may have an affect as the participants may answer dishonestly. They also rely on retrospective data, which can be inaccurately recalled.
• Association does not mean causality. There may be factors other than the internal working model that influence later relationships, such as parenting style.
• Bowlby and other psychologists may have exaggerated the influence of the internal working model on adult relationships. It is a very deterministic view that suggests people who had attachment problems in childhood would not be able to form good relationships in the future.