Attachment theory and cognitive models of depression have each identified mechanisms that contribute to emotional vulnerability, yet vast parts of the intersection of these frameworks remains underexplored. This project investigates how attachment security—both as a trait and as a state manipulated through priming—modulates inferential processes that underlie depressive symptoms. Specifically, we examine how secure attachment (dispositional or induced) influences the inferences made for personal negative events as well as shifting from an initial depressogenic inference towards a more adaptive one. Across four studies, we employ experimental, diary, and cognitive training paradigms to test whether enhancing attachment security reduces inferential negativity, promotes adaptive inferential shifting, and lowers rumination and negative mood. Study 1 tests whether secure attachment priming influences inferences about recalled romantic conflicts. Study 2 uses a diary study to assess the impact of such priming on inferences made for daily negative events. Study 3 assess whether priming enhances inferential flexibility for feared and real-life negative events, respectively. Study 4 evaluates whether combining attachment priming with cognitive bias modification (CBM) improves intervention outcomes. This multi-method program aims to advance an integrated cognitive-interpersonal model of depression vulnerability and inform the development of more personalized and process-focused interventions. By linking attachment dynamics to modifiable cognitive mechanisms, the project offers novel insights into how emotional bonds shape meaning-making and psychological resilience.