Understanding the impact of forgetting is crucial in neuronal systems, particularly in how it affects the remodeling of neuronal circuitry. In this study we explore the role of forgetting in the context of reversal learning in mice, complemented by a simulated analysis of synaptic remodeling in a recurrent neural network.In our experiment, mice were trained to discriminate between two stimuli in a go/no-go task, followed by a retention test after intervals of either 2 or 16 days. This phase was succeeded by reversal learning, where the stimulus contingencies were switched. We observed that mice with the shorter pause exhibited not only enhanced performance in the memory test but also faster reversal learning, indicating a preserved memory of the task structure.To contextualize these findings, we employed a single-layer recurrent neural network successfully trained on a binary classification task. This initial learning was followed up by reversal learning as in the experiment. As a theoretical model for forgetting, we progressively randomized fractions of synaptic weights in the network’s recurrent connections. This approach correlated with our experimental data, revealing an inverse relationship between the degree of synaptic weight randomization and the pace of reversal learning.Conclusively, our research illuminates the intricate relationship between forgetting, synaptic remodeling, and reversal learning. The observed delay in reversal learning after an extended break post-initial learning phase implies a gradual diminishment in the memory's capacity to retain task-relevant structures over time.