Pharmacological agents theorized to modulate fear extinction could enhance treatments for anxiety and trauma-related disorders, but fear conditioning and treatment studies testing these agents often yield null or conflicting results. We review principles of extinction learning relevant to the design of studies that test pharmacological enhancements of extinction. We then critically review the methodologies of existing studies for three pharmacological agents [d-cycloserine (DCS), glucocorticoids (GCs), and L-DOPA] with respect to key learning principles. While each agent has promising support in rodent models, many human studies are not designed to adequately detect an agent’s effects on extinction learning. We provide specific recommendations, informed by these learning principles, for future study designs that may clarify whether, how, and under what conditions agents impact extinction.