<p>One-trial learning is a critical evolutionary mechanism that enables animals and humans to remember important events rapidly and adapt to dynamic changes in the environment. This article reviews a variety of one-trial learning paradigms, including passive avoidance, taste aversion, conditioned freezing, food association, object place recognition, and social learning. Experimental models of one-trial learning span many animal species and vary in ecological validity, sensory modality, type of conditioned stimuli, valence of reinforcement, and the pattern of brain structures involved in learning in different models, providing for studies of a wide range of cognitive processes. The rapidity of memory formation in these paradigms provides unique opportunities to control experimental variables and minimize artifacts. Moreover, the clear isolation of the learning moment in these models makes them indispensable for studying the phases of memory consolidation and their neural mechanisms – especially neural and synaptic plasticity – at the earliest stages of this process. These points make one-trial learning models a powerful tool for studying the fundamental neurobiological mechanisms of memory and learning, and they can also contribute to the development of new approaches to the treatment of cognitive disorders associated with memory impairments, such as amnesia and post-traumatic stress disorder. Furthermore, the patterns revealed using these approaches may provide the basis for the development of fast learning algorithms for artificial intelligence.</p>