Although we experience everyday life as a continuous stream of information, we generally perceive and remember this information as more discretized episodes. This type of memory, termed episodic memory, involves integrating information into an episode and associating it with the spatiotemporal context in which it occurred. Episodic memories do not occur in isolation, but rather may be associated to one another based on their shared features. As such, characterizing episodic memory processes can be particularly challenging, despite the critical role of episodic memories to our everyday lives. In this talk, Lohnas will present a computational model she developed that formalizes the cognitive processes underlying episodic memory formation, storage and retrieval. This model assumes that the shared contextual features amongst episodic memories are critical to how they are represented and retrieved. In a series of simulations, Lohnas will demonstrate that this model can account for established memory findings in the episodic memory paradigm of free recall. Further, Lohnas will present data testing several novel predictions of the model, as manifest in behavior and neural data. All of the novel model predictions were upheld, and are challenging to explain without the model’s assumptions of context. Combined with the model’s ability to naturally account for existing free recall effects, the results converge on a strong role for context in the organization of episodic memories. Lohnas will end by presenting neural data that can be used to inform future modeling work.