Prenatal Attachment Is Associated with Attachment-Related Representations at Child Age 30 Months Among Young Mothers

Background and Purpose: During pregnancy, women develop thoughts and feelings about their unborn children that attachment theory would suggest form the basis of the mother’s later relationship with the child.  A small body of research has shown associations between mothers’ perceptions of their unborn children and the ways they interpret their newborns’ behavior (Zeanah et al., 1990; Rubin, 1984), as well as the affective tone of their postnatal representations of their infants and their sensitive caregiving of their older infants (Dayton et al., 2010; Crawford & Benoit, 2009; Theran et al., 2005). However, no studies have examined associations between prenatal maternal attachment and representations or parenting beyond infancy.  The purpose of this study is to examine whether prenatal attachment is predictive of maternal attachment representations of their toddlers among young mothers.

Method:  The sample includes 182 women who participated in a longitudinal study from pregnancy until their children were 30 months of age.  The women were young (M=18.4 years), low income, and from diverse racial/ethnic backgrounds (45% African American, 38% Latina, 17% multiracial/other).  At 37 weeks of pregnancy the women completed the Maternal Antenatal Attachment Scale (MAAS: Condon, 1993), a structured questionnaire measuring the mother’s quality of attachment and preoccupation with her unborn baby.  At child age 30 months, the mothers participated in the Working Model of the Child Interview (WMC: Zeanah et al, 1986), a semi-structured interview probing parents’ attachment-related representations of their young children.  The WMC was audio-recorded, transcribed, and coded by coders who were trained and reliable.  Coding included rating themes in the mothers’ interviews, and also categorization of the narratives as balanced, distorted or disengaged. Multinomial logistic regression was conducted to examine the association between prenatal attachment and 30 month attachment categories and ratings.

Results:  At 30 months postpartum, 61% (n=111) of mothers had balanced representations of their child, 25% (n=46) had distorted representations, and 14% (n=25) had disengaged representations.  Regression analyses showed that mothers classified as balanced had higher prenatal attachment scores than both the disengaged mothers (p<.05) and the distorted mothers (p<.01).  Prenatal attachment was also positively associated with WMC ratings of acceptance (p<.001) and openness (p<.01), and negatively associated with helplessness (p<.001) and difficulty (p<.01).  Mother demographics measured during pregnancy, including age, relationship with the baby’s father, and co-residence with her parent figure were not predictive of either prenatal attachment or WMC classification. 

Conclusion: Results of this study demonstrate that how a mother thinks and feels about her unborn baby is strongly connected to her thoughts and feelings about the same child several years later. Mothers with distorted or disengaged representations of their 30 month child reported lower quality of attachment and were less preoccupied with thoughts of their baby during pregnancy.  The prenatal period may be an important time to implement interventions that help mothers develop an emotional bond with their baby, recognize their baby’s needs and capabilities, and explore reasons why they may be struggling to connect with their childre