Stability of attachment-related anxiety and avoidance and their relationships with the five-factor model and the psychobiological model of personality

Objectives. The stability of attachment-related anxiety and avoidance and their relationship with influential personality models are issues worth further investigation. Studying discriminant validity with respect to aspects of personality deemed as prevalently heritable, such as temperament, is particularly important. Our aims were to examine the stability of attachment dimensions and to study their relationship with psychological distress, the Big Five factors, and temperament and character dimensions. Methods. The Experiences in Close Relationships (ECR) questionnaire, the State-Trait Anxiety Inventory (STAI), the Zung depression scale (ZDS), the Temperament and Character Inventory (TCI-125), and the Big Five Questionnaire (BFQ) were administered to 222 undergraduate and graduate students. After 1 month, 115 participants completed again the ECR, the STAI, and the ZDS. Results. Attachment-related anxiety was modestly correlated with depression and anxiety, while attachment-related avoidance was modestly correlated with depression. The test-retest reliability of ECR scores was high. Changes in attachment-related anxiety were not related to changes in depression or anxiety, and changes in attachment-related avoidance were modestly correlated with changes in depression. In separate multiple regression models also including gender and age, attachment-related anxiety was correlated with harm avoidance, reward dependence, low novelty seeking, low self-directedness and low cooperativeness; low energy/extraversion and low emotional stability; attachment-related avoidance was correlated with low reward dependence and tended to be correlated with low self-directedness. Conclusions. These findings corroborate the reliability and the convergent and discriminant validity of the ECR, and support the notion that the attachment dimensions are only modestly related to psychological distress and are not redundant with constructs developed within influential personality models such as the five-factor model and the psychobiological model