Attachment and representational models

Reminiscent of object representations of self and others from Object Relations theory and, later, Stern’s RIGS (representations internalized and generalized), Bowlby described three aspects of Attachment Theory: “Internal Working Models,” attachment behaviors (crying, clinging, proximity, etc) and ‘defensive exclusion’ [denial, dissociation, etc]. Bowlby recognized that the environment (i.e. relationships) influenced development and that healthy attachments helped a child weather stressful life events. What is to be emphasized is it is the relationship with its implicit relational knowing that is ‘internalized’ (encoded in the brain). Because it is the relationship (not simply the ‘object representation’) which is encoded, either pole [or both] of that relationship (e.g. victim- rescuer; doer-done to, etc) is accessible to be enacted.  

While Object Relations theory writes about unstable representations of self or other, it might sometimes be better conceived to think in terms of self states, or multiple selves, each state with its ‘stable’ [rigid] adherence to a particular way of being with a particular attachment figure, also called implicit relational knowledge. ‘Instability’ or switching of self states, with other states moving to the background or dissociated, including its concomitant affective dysregulation, is the hallmark of [relational] trauma, often attributed to personality disorders with ‘borderline’ organization. [Because incoherence in narrative -- see Adult Attachment Interview--, lability of affect, denial and dissociation all make a person look ‘borderline psychotic’].

What research now reveals is that insecure attachment, especially later preoccupied and/or unresolved, is predominant in Borderline Personality Disorder (BPD), and that emotional violence/neglect are more strongly correlated with development of BPD than even the trauma of sexual abuse.

Buchheim, A. and Kächele, H. (2003). Adult Attachment Interview and Psychoanalytic Perspective. Psychoanal. Inq., 23(1):81-101