Attachment styles and close relationships: A four‐year prospective study

Abstract A longitudinal study of 177 adults examined the stability of adult attachment styles and of romantic relationships over a 4‐year period. Findings included the following: (a) attachment styles were highly stable over time; (b) Time 1 attachment style was a significant predictor of Time 2 relationship status, but (c) this effect was mediated by concurrent attachment style at Time 2; (d) secure respondents were less likely than insecure respondents to report one or more breakups during the 4‐year interval, but (e) paradoxically, ambivalent respondents were just as likely as secure respondents to be in a relationship with the same partner they had identified 4 years earlier; and (f) attachment stability was moderated to some extent by the experience of breakup or initiation of new relationships during the interim. Respondents' ability to recall their previous attachment style was also examined. Methodological and theoretical implications are discussed, particularly with respect to the conceptualization of attachment styles as traits versus reflections of current relationship status.