Making optimal decisions depends on appreciation of the value of choices. An important source of information about the value of choices is prior experience: Positive experiences are more likely than negative ones to be selected again. Such value-based learning has long been considered the domain of trial-and-error incremental learning mediated by a cortico-striatal system. Recent evidence, however, suggests that in some conditions the medial temporal lobe (MTL), which mediates episodic memory, may also contribute to value-based learning. However, when and how the MTL contributes is poorly understood. In this proposal we put forth a novel way to conceptualize the role of the MTL in decision-making that exploits the known computational characteristics of that system. Specifically, MTL-mediated episodic memory representations are flexible and serve not only memory but also the representation of possible novel scenarios. We postulate that the representational flexibility afforded by the MTL is important for decision-making whenever value information needs to be integrated across distinct experiences. Such is the case when 1) value information needs to be updated as a result of shifting reward contingencies; 2) when value information is not directly available but is generalized across experiences; and 3) when value representations need to be constructed de novo. We systematically examine the performance of patients with MTL lesions in these three paradigmatic cases and test specific predictions about the conditions in which value-based learning and decision-making will be impaired. The studies comprising this proposal have the potential to deepen our understanding of the role of the MTL in higher-level cognition and to break down the stark boundaries commonly assumed between processes involved in episodic memory and decision-making. Further, it is anticipated that these studies will have important clinical implications for a range of neurological and psychiatric populations with MTL pathology. Understanding the scope and nature of MTL involvement in decision-making not only will shed light on the nature of decisional impairments in these populations but also will guide clinical assessment and theoretically informed interventions.