 The hippocampus plays an important role in learning and memory consolidation through replay of neuronal sequences. These sequences are constructed from past experiences and can be modified according to current spatial constraints. However, replay does not always reflect previous behaviors or construct new ones. A new stochastic replay mechanism has been proposed which prioritizes experiences based on three factors, experience strength, experience similarity, and inhibition of return. This prioritization results in more efficient replay and improved performance compared to random replay. Additionally, the model is able to reproduce diverse replay statistics due to its stochastic nature and the fact that different experiences result in different weights for each factor. This article was authored by Nicholas Dykeman and Sen Cheng.