The association between attachment and psychopathic traits

Abstract

The purpose of this study was to examine the association between attachment and psychopathic traits. A systematic search of relevant articles yielded 12 studies (11 independent samples), containing 133 effect sizes based on in total 1876 participants. Results from a three-level random-effects meta-analysis demonstrated a small-to-medium significant, overall association of r = 0.18, indicating that insecure attachment was positively related to psychopathic traits. We also tested moderator effects; six significant moderators were found. First, the magnitude of the effect size depended on which psychopathic trait was investigated; the largest effect size was found for callous-unemotional traits (r = 0.23), and no effect was found for antisocial behavior, narcissism and impulsivity. Second, parent- and caregiver reports on psychopathic traits resulted in stronger associations compared to self-reports and a composite of multiple sources. Third and fourth, the association varied across attachment and psychopathic traits instruments. Fifth, if both attachment and psychopathic traits were reported by the same informant, the association was stronger than when attachment and psychopathic traits were reported by different informants. Sixth, the association differed across sample setting, with the strongest association for a clinical setting followed by a secured setting and no association for both community or combined settings. No moderating effects were found for the attachment categories, attachment to whom and the informant of attachment, publication year, and sample characteristics, such as gender, age, and sample size. We discuss the implications of our findings for future research and clinical practice.