Relations between maternal depression and attachment security among 50 infant-mother and 54 preschool child-mother dyads were examined using the classification system of M. D. S. Ainsworth, M. C. Blehar, E. Waters, and S. Wall (1978) and M. Main and J. Solomon (1990) for infants and the Preschool Assessment of Attachment (P. M. Crittenden, 1992b) for preschoolers. Attachment insecurity was significantly associated with maternal depression among infants and preschoolers. Furthermore, children without unitary, coherent attachment strategies tended to have more chronically impaired mothers than did children with coherent, organized attachment strategies. Results stress the importance of severity-chronicity of parental illness in the study of depression and early attachment relations, and that differences between children with and without coherent, organized attachment strategies are as clinically informative as are differences between secure and insecure children.

This study sought to replicate previous work in testing the hypothesis that interactions of dyads developing secure attachment relationships would be characterized by disproportionately synchronous and those of dyads developing insecure relationships by disproportionately asynchronous exchanges. Additionally, a priori hypotheses were tested regarding expected differences in the interactional histories of dyads developing insecure-avoidant and insecure-resistant attachments. Results supported the study's predictions in all cases. Dyads developing secure attachments were observed at 3 and 9 months to interact in a disproportionately well-timed, reciprocal, and mutually rewarding manner; dyads developing insecure relationships were disproportionately characterized by interactions in which mothers were minimally involved, unresponsive to infant signals, or intrusive. Within the insecure group, as predicted, 3- and 9-month interactions of avoidant dyads were characterized by maternal intrusiveness and overstimulation; resistant dyads were characterized at both ages by poorly coordinated interactions in which mothers were underinvolved and inconsistent. These findings are discussed as they lend to a growing body of evidence concerning associations between differential interactional histories and attachment quality.

Observed 1-, 3-, and 9-month infant-mother interaction to examine antecedents of 1-year attachment quality. Frequency data were recoded using theory-guided measure of interactional synchrony; chi-square and prediction analyses tested hypothesis that development of secure attachments is predictable from synchronous, and insecure attachments from asynchronous interactions across first year. Findings from 30 dyads < 10 secure, 10 avoidant, 10 resistant) supported hypothesis at I and 3 months, with synchronous interaction observed at significantly, disproportionately frequent rate for securely attached dyads. Also identified theoretically consistent aspects of interaction (e.g., responsiveness) that differentiated mothers of secure, avoidant, and resistant babies. Authors discuss findings as they support major tenets of attachment theory and suggest usefulness of a priori methodological approach.12 This document is copyrighted by the American Psychological Association or one of its allied publishers.This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

This study was based on the premise that much of the instability evident in research on infant emotionality/temperament is a function not so much of measurement error (as typically presumed) but lawful discontinuity. Infants who changed from high to low and from low to high levels of negative or positive emotionality between 3 and 9 months of age were compared with infants who remained stable during the period on distal measures of the family environment (prenatally and neonatally measured) and proximal measures of parent-infant interaction (3 months) thought to account for stability and change in infant emotionality and on 1-year infant-mother attachment security. Results revealed more success in forecasting stability and change in negative emotionality than positive emotionality; maternal personality and marital factors and mother-infant interaction accounted for why infants highly negative at 3 months changed, and comparable father factors and processes accounted for why infants initially low in negativity changed. Attachment-related analyses revealed change in positive emotionality to be more related to 1-year security than change in negative emotionality, but it was also the case that continuity and discontinuity in both positivity and negativity interacted to forecast attachment security.

This study built on attachment theory and previous research in examining the interactional origins of the secure, insecure-resistant, and insecure-avoidant patterns of attachment. Maternal sensitive responsivity, rejection, and activity were the focus of repeated naturalistic observations when infants were 1, 4, and 9 months of age; quality of attachment was assessed at 1 year. Mothers of secure 1-year-olds were observed to be more sensitively responsive at 1 and 4 months and less rejecting at 1 and 9 months than mothers of insecure infants. Mothers of insecure-avoidant infants were more rejecting at 9 months, whereas mothers of insecure-resistant infants were least sensitively responsive and most rejecting at 1 month; the insecure groups were also differentiated on the basis of patterns of change from 1 to 9 months, with mothers of resistant infants becoming less rejecting and mothers of avoidant infants becoming more rejecting relative to other mothers.