Dimensional Attachment in Romantic Relationships: Its Cognitive–Evolutionary Definition and Possible Research Implications

Background. Attachment between romantic partners has been investigated primarily through self-reports assessing the two dimensions of avoidance and ambivalence/anxiety, typically formulating items according to observable psychological properties. However, this approach produces correlated dimensions, with a limited distinction between their features. In this work, we hypothesize that a definition of avoidance and ambivalence driven by their theorized cognitive–evolutionary functions can lead to a clearer-cut distinction between the two (i.e., lower correlation), still maintaining or even further highlighting the expected avoidant and ambivalent characteristics. Methods. To test our hypothesis, we administered to an adult population a battery of self-reports measuring attachment and related properties—i.e., psychological well-being, couple satisfaction, and emotion regulation. We assessed attachment through two classical attachment self-reports—the Experiences in Close Relationships-Revised (ECR-R) and Revised Adult Attachment Scale (RAAS)—and the novel, theory-informed Attachment–Caregiving Questionnaire (ACQ), building a Partial Least Squares Structural Equation Model (PLS-SEM) connecting attachment with the other constructs. Results. Our results demonstrate that measuring attachment through the ECR-R/RAAS and the ACQ leads to similar properties but a more evident characterization of avoidance and ambivalence supported by the ACQ. Conclusions. These first findings suggest that the cognitive–evolutionary definition of these two dimensions may be advantageous and should be further investigated.