Schaffer's stages of attachment

• Stage 1 = asocial stage (first few weeks)

  • Baby’s behaviour towards people and inanimate objects is quite similar
  • Some preference for familiar people (more easily calmed by them)
  • Babies are also happier in the presence of other people
• stage 2 = indiscriminate attachment (2-7 months)

  • Babies now display more observable social behaviour with a preference for people rather than inanimate objects
  • They recognise and prefer familiar people
  • Babies do not show stranger or separation anxiety
  • Attachment is indiscriminate because its the same towards all
• stage 3 = specific attachment (from ~7 months)

  • Stranger anxiety and separation anxiety when separated from one particular person
  • Baby is said to have formed a specific attachment with the primary attachment figure
  • This is in most cases the person who offers the most interaction and responds to the baby’s ‘signals’ with the most skill (the mother in 65% of cases)
• stage 4 = multiple attachments (by one year)

  • Secondary attachments with other adults form shortly after
  • Schaffer and Emerson’s study = 29% of babies had secondary (multiple) attachments within a month of forming a primary (specific) attachment
  • By the age of one year = the majority of infants had multiple secondary attachments
• Schaffer and Emerson, stages of attachment study (procedure)

  • 60 babies from Glasgow (most from working-class families)
  • Researchers visited babies and mothers at home every month for a year and again at 18 months
  • Separation anxiety = measured by asking mothers about their children’s behaviour during everyday separations (eg adult leaving the room)
  • Stranger anxiety = measured by asking mothers questions about their children's anxiety response to unfamiliar adults
• Schaffer and Emerson, stages of attachment study (findings/conclusions)

  • Babies developed attachments throughout a sequence of stages - from asocial through to a specific attachment to multiple attachments
  • The specific attachment tended to be the person who was most interactive and sensitives to babies‘ signals and facial expressions (ie reciprocity)
  • This was not necessarily the person the baby spent most time with
• strength = S+E study has external validity

  • Most of the observations (not stranger anxiety) were made by parents during ordinary activities and reported to researchers
  • The alternative would be to have observers present in the babies‘ homes - this may have distracted the babies or made them feel more anxious
  • This means it is highly likely that participants behaved naturally while being observed
• counterpoint to study having external validity

  • mothers may have been biased in what they reported
  • Eg = they might not have noticed their baby was showing signs of anxiety or may have misremembered it
  • Means that even if babies behaved naturally, their behaviour mat not have been accurately recorded
• limitation = poor evidence for the asocial stage

  • Because of their stage of physical development, young babies have poor coordination and are fairly immobile
  • This makes it difficult for mothers to accurately report signs of anxiety and attachment for this age group
  • Means the babies might actually be quite social but, because of flawed methods, they appear to be asocial
• strength = RWA to day care

  • In early stages (asocial and indiscriminate attachments) babies can be comforted by any skilled adult
  • but if a child starts day care later (during the stage of specific attachments) care from an unfamiliar adult may cause distress and longer-term problems
  • Means that S+E stages can help parents making day care decisions
• extra evaluation = generalisability

  • S+E based their stages on a single but large-scale study of babies‘ development conducted in working-class Glasgow
  • BUT = child-rearing practices vary considerably according to cultural and historical context
  • eg = van ijzendoorn = multiple attachments the norm in collectivist cultures
  • Means that some of the observations from this study may not generalise to other populations