Attachment security in children with autism

Lisa Capps, Marian Sigman, Peter Clive Mundy

Research output: Contribution to journalArticlepeer-review

124 Scopus citations


Nineteen autistic children were examined in a modified version of Ainsworth's Strange Situation. The attachment security of 15 children could be classified. Each of these children displayed disorganized attachment patterns, but almost half (40%) of them were subclassified as securely attached. To assess the validity of the attachment classifications, children and their mothers were observed in a separate interaction. Mothers of children who were subclassified as securely attached displayed greater sensitivity than mothers of children who were subclassified as insecurely attached. Children who were subclassified as securely attached more frequently initiated social interaction with their mothers than did children who were subclassified as insecurely attached. Children with secure and insecure subclassifications were compared to investigate correlations between attachment organization and representational ability and social-emotional understanding. Although children with underlying secure attachments were no more likely to initiate joint attention, they were more responsive to bids for joint attention, made requests more frequently, and demonstrated greater receptive language ability than children subclassified as insecurely attached. Discussion focuses on dynamics that may contribute to individual differences in the attachment organization of autistic children and on the reciprocal relationship between advances in our understanding of normal and pathological development.

Original languageEnglish (US)
Pages (from-to)249-261
Number of pages13
JournalDevelopment and Psychopathology
Issue number2
StatePublished - 1994
Externally publishedYes

ASJC Scopus subject areas

  • Developmental and Educational Psychology
  • Psychiatry and Mental health


Dive into the research topics of 'Attachment security in children with autism'. Together they form a unique fingerprint.

Cite this