Not just mothers: Understanding the unique role of attachment to fathers

Attachment theory (Bowlby, 1969) is one of the most influential theories regarding the impact of family relationships on lifelong relational and individual well-being. Through the lens of attachment theory, patterns of interactions with parents (and other important figures) can shape working models of the self and others which persist into adulthood. When parents are absent or unreliable, children may develop anxious attachment patterns (where they may be overly afraid of rejection, abandonment, and withdrawal), and/or avoidant attachment patterns (where they may withdraw, avoid intimacy, and distrust others). On the other hand, children who have reliable, supportive parents should develop a “secure” attachment and feel more comfortable and confident in exploring the world around them. Attachment theory suggests that both attachment anxiety and attachment avoidance can be detrimental to psychological well-being. This hypothesis been supported by a large body of empirical research in children, adolescents, and adults suggesting a strong link between attachment and mental health (e.g., Madigan et al., 2016; Spruit et al., 2020; Zheng et al., 2020).

Although attachment theory has motivated research and shaped practice for decades, the vast majority of empirical work has focused on the role of mothers to the exclusion of other caretakers. Most notably, research on attachment to fathers remains highly limited (Bretherton, 2010; Brumariu & Kerns, 2010). This has led to limited understanding of how, if at all, attachment relationships with fathers differ from those with mothers, and what the combined impact of these relationships is on development. These criticisms are not new (e.g., Cowan, 1997; Rutter, 1979), but recent increases in research on attachment to fathers may signal that the tides are changing (Ahnert & Schoppe-Sullivan, 2020). What might we expect to learn from this work as it develops?

Some of the most important questions regarding paternal attachment may be comparative: how do the effects of paternal attachment stack up to those of maternal attachment, and are these attachment relationships interdependent? Van Ijzendoorn and colleagues (1992) proposed four conceptual “multiple-caregiver” models which may help us consider the possibilities.

First, there is the monotropy model, which suggests that only attachment to the primary caregiver (i.e., the mother) is truly relevant for well-being. This idea harkens back to the foundations of attachment theory; while Bowlby acknowledged that other relationships could be important, the concept of a monotropic, maternal relationship shaped his work on maternal deprivation. The maternal deprivation hypothesis suggests that, absent a secure relationship with the mother during the formative period, children will develop serious emotional, intellectual, behavioral, and social problems.

Second, there is the hierarchical model, a softer version of monotropy. In this model, there may be important contributions of secondary caregivers that are worth considering, even as the primary caregiver reigns supreme. For example, perhaps the maternal relationship remains the most important for all aspects of development, but strong relationships with fathers and other caregivers could buffer children from the worst effects of maternal deprivation.

Third, there is the independence model, which suggests that attachment relationships may have qualitatively different, rather than quantitatively different, effects. For example, perhaps the maternal attachment relationship is more important for developing social-emotional skills, whereas the paternal attachment relationship is more important for developing a sense of autonomy. Because many different skills and internal processes are necessary for well-being, the independence model would suggest that these different effects may not be as directly comparable, but should be understood within their own contexts.

Finally, there is the integration model, which suggests that attachment relationships are part of a system that should be considered more holistically. That is, the true impact of maternal attachment can only be understood alongside the context of other attachment figures. For example, having an anxious maternal attachment may have a different meaning and impact depending on the child’s relationship with their father or other caregivers. In this view, to understand relative risk in developing mental health problems, for instance, it would be necessary to consider these different attachments and how they may affect each other rather than simply adding them up as separate risk factors.

To explore some of these possibilities, we have collected some preliminary data from a residential mental health treatment center for adolescents, the Newport Institute. The Newport Institute has been collecting systematic data, in collaboration with the Drexel University Center for Family Intervention Science (CFIS) research team, to improve quality of services for adolescents and young adults in their treatment centers. Data are collected at intake and at regular follow-up intervals. Assessment systems are designed and analyzed by the CFIS team led by Guy Diamond, PhD, and including Linda Ruan-Iu, PhD; Payne Winston-Lindeboom, PhD; and the author of this blog post.  

This project examined self-reported attachment at intake to both mothers and fathers separately (Rivers et al., 2021). These data generally suggest that attachment to both parents separately is uniquely related to symptom severity for anxiety and depression. Moreover, analyses using response surface analysis suggest some support for the integration model. Specifically, for some combinations of symptoms and attachment dimensions, maternal and paternal attachment may interact and even demonstrate “mismatch” effects, where having different attachment patterns across parents is associated with more severe symptoms. However, maternal attachment was still the primary predictor in most models, and the most robust after controlling for covariates.

Research on paternal attachment is still developing. Although more work needs to be done to determine which conceptual models may best fit the reality of paternal attachment, recent research suggests that the monotropic model is insufficient. For example, one meta-analysis found similar effect sizes for maternal and paternal attachment in relation to internalizing symptoms (Deneault et al., 2021). As this research develops, it should have important implications for clinical interventions that have their roots in attachment theory, such as Attachment-Based Family Therapy (Diamond et al., 2013), Attachment and Biobehavioral Catch-up (Dozier & Bernard, 2019), connect (Moretti & Obsuth, 2009), and Circle of Security (Powell et al., 2016). The role of fathers in these types of interventions will need to be carefully considered in light of this developing line of work. As families become more complex, attachment research needs to catch up to the reality of multiple caregivers, including the possibilities of multiple people taking on mother and father roles in blended, multigenerational, and LGBTQ+ families. Given the recognized importance of these attachment relationships for healthy development, these investigations can have a real impact on the lives of children, adolescents, and young adults.