Anxiety disorders are the most common mental disorders and are often chronic and disabling. Although exposure-based treatments are effective, a substantial number of individuals fail to fully remit or experience a return of symptoms after treatment. Understanding the critical processes underlying the development and treatment of anxiety disorders will help identify individuals at risk and optimize treatments. Aversive associative learning offers explanatory pathways through which fear and anxiety emerge, spread, persist, and resurge. This narrative review examines the advances made in our understanding of associative fear and avoidance learning in anxiety disorders. Overall, the extant literature supports a key role of aversive associative learning in the development and treatment of anxiety disorders. However, research targeting specific mechanisms such as extinction generalization and avoidance, the fragility of extinction, and moderating influences of individual differences pertinent to anxiety disorders (e.g., age, sex, depression) is needed. We discuss the need for more ecological valid and complex paradigms to model ambiguity and conflict as well as for clinical translation studies to optimize treatment.