Testing the stability of adult attachment in romantic relationships from an evolutionary perspective

Abstract

According to attachment theory, individuals who were secure as a child will grow up to be secure in their romantic relationships. This is because, once internal working models of attachment are formed, they remain relatively stable across the lifetime. The aim of this thesis was to test the stability of adult attachment in romantic relationships from an evolutionary perspective. Five studies were conducted in this endeavour, each addressing the stability of attachment style in romantic relationships in response to either varying socio-ecological conditions or physiological state. The rationale of the thesis is that these or similar variations might have been recurrently experienced over the course of human evolution and thus are likely to be potential triggers for adaptive responses. The first three studies tested the stability of attachment and conditional changes in the short term, while the remainder investigated more long-term changes. In the first three studies, participants’ attachment security was measured twice, before and after a manipulative condition. In study 1, participants’ self-perceived mate value was manipulated by showing them a series of either same-sex attractive, unattractive or a mixture of attractive and unattractive images. In study 2, participants’ perceived operational sex ratio was manipulated by showing them either female-biased, male-biased or equal ratio of female and male faces. In study 3, participants were exposed to subtle relationship breakdown cues. The results of these studies showed that individuals’ partner attachment security could be changed in the short-term, depending on current experience and the perceived social environment. In study 4, to test the stability and change depending on long-term physiological changes, I investigated whether congruent oral contraceptive use influences partner attachment security. Congruency in oral contraceptive use was found to influence the partner attachment security in coupled women, such that previous and current OC users reported higher partner attachment anxiety. The studies also revealed that the origin of influences on general attachment security in relationships is not entirely parental: attachment to partners is a better predictor of attachment in romantic relationships. Study 5 provided cross-cultural support for this by comparing results from individualist and collectivist countries. In conclusion, adult attachment is flexible and could be changed depending on the current experiences. Attachment is multifactorial and complex, and is not only associated with childhood experiences.