The Effects of Sleep and Reinforcement on the Formation of Transitive Inferences and Relational Memory

Abstract

Recent research suggests that dopamine facilitates the formation of long term memory and may allow for the adaptive use of memory. During reinforcement learning, dopamine is released in the ventral tegmental area of the brain (VTA), which facilitates long term potentiation (LTP). The VTA is also reactivated during sleep, although the function of this reactivation is currently unknown. Additionally, there is evidence that sleep also facilitates the generalization of memory; however an interaction between the dopaminergic system and sleep on the adaptive use of memory has not been directly investigated. Thus, in this study we investigated the combined effects of the interaction between sleep and reinforcement during learning on performance in a transitive inference task, which requires the integration and generalization of disparate premise memories. Our results showed that: 1) reinforcement during learning is required for successful generalization in transitive inference; 2) when reinforcement occurs during learning, sleep further enhances the ability to make transitive inferences, compared to an equal period of wake. These data provide evidence of an interaction between sleep and the dopamine reward system to facilitate generalization from existing memories.