Computational model of sleep mediated memory consolidation
Sleep. Memory. Computational Models. Machine Learning.
The brain stores memories by persistently changing the conectivity between neurons. Sleep is known to be critical for these changes to endure. Researches in the neurobiology of sleep and the mechanisms of long-term synaptic plasticity have provided data in support of several theories of how the sleep affects the long-term synaptic plasticity. The existing theories, although supported by experimental data, are sometimes contradictory, with some evidences pointing to the role of sleep in the forgetting of irrelevant memories, whereas other results indicate that sleep supports the reinforcement of the most valuable recollections. Computational modelos and simulates provide basis to quantitative tests and comparisons between theory predicts and observed data, and might also work as strategy to organize a pool of data and methodologies in sleep research. This work outlines the emerging progress in the computational modeling and simulation of the main theories on the role of sleep in memory consolidation. Most theories have been backed up by computational evidence, but the different studies show little integration. This research proposes a computational framework to conciliate different views about the role of sleep in the memory consolidation. The work also intends to improve the capacity of smart learning systems, using principles from neurobiology of sleep.