The Pennsylvania Infant and Family Development Project, III: The Origins of Individual Differences in Infant-Mother Attachment: Maternal and Infant Contributions

In order to test 4 specific hypotheses regarding the interactional histories associated with variation in quality of infant-mother attachment, data gathered during naturalistic home observations at 1, 3, and 9 months on 60 babies seen in the Ainsworth and Wittig strange situation were examined. Planned comparisons revealed, as predicted, that securely attached infants had experienced intermediate levels of reciprocal interaction and maternal stimulation, considered to be reflective of sensitive care, and that resistant babies had experienced less responsive care than securely attached infants. No support was provided for the hypothesis that avoidant babies had experienced less physical contact with mothers than securely attached infants. Insecurely attached infants were observed to cry significantly more than securely attached infants at 3 and 9 months. A cross-lag panel analysis, designed to assess longitudinal processes of influence, revealed that fussiness was caused by mothering and did not serve to influence mothering. These results are discussed in terms of mothers' relatively greater influence in determining individual differences in attachment, with overstimulation leading to avoidance, understimulation leading to resistance, and intermediate levels of stimulation leading to security.