View article

The concept of mental working models that are relationally based, affectively colored, dynamic, and integrative is one of the most generative concepts of attachment theory. It is also one of its most difficult. The generativity of this concept is reflected in how important it is to theoretical explanations for the association between the security of attachment and its correlates, the stability of security over time, and intergenerational consistency in attachment. Attachment researchers frequently use internal working models (IWMs) as their explanation for research findings. In light of these applications, it seems surprising that Bowlby did not devote greater attention to elucidating the concept of IWMs in his writings, perhaps leaving this to his followers. And therein lies the difficulty. Without a well-developed theoretical account of the nature and development of IWMs to guide their inquiry, attachment researchers have created a variety of conceptualizations to fit different theoretical and empirical needs. More than 30 years ago, Hinde (1988) stated the problem succinctly:“In the very power of such a model lies a trap: it can too easily explain anything”(p. 378), and since then IWMs have been enlisted to “explain” the association of attachment with political ideology, math achievement, and many other phenomena (see Thompson, 2016). Attachment theory has not been well served by a core theoretical concept that is so vaguely defined and flexibly applied.