Studies on animals have shown that the neuropeptide oxytocin is involved in maternal behaviour and pair-bond formation (Insel, Young, 2001). It also seems plausible that oxytocin plays a key role in adult attachment relationships and prosocial behaviour in humans (Heinrichs, Domes, 2008). Intranasal administration of oxytocin demonstrated beneficial effects on social cognition (e.g. improvement in emotions detection) and stress regulation (e.g. alleviating HPA axis; for a review, see Meyer-Linderberg et al., 2011). However, initial optimism about main effects has been questioned and a nuanced view about oxytocin has been introduced regarding context and/or person as a moderator (Bartz et al., 2011).

In this poster, we further elaborate this interactionist approach from an attachment theory perspective. We outline Mikulincer & Shaver's (2007) multi-modular model of attachment system-activation and extend this view by evidence that individuals posses multiple attachment relationships. In line with this, we underline researchers' current attention for general and relationship-specific internal working models of attachment (IWMs; Fraley et al., 2011).

In this context, we review 4 published papers on the topic and their experiments in normal controls and patients with borderline personality disorder regarding the differences between general and specific IWMs by emphasizing the kind of methods used by researchers and their results. Therefore, we outline future hypotheses to be investigated and implications of this approach for further research in order to explain inconsistent and sometimes unexpected findings of the effects of oxytocin in the social domain.