PT  - JOURNAL ARTICLE
AU  - Oscar C. González
AU  - Yury Sokolov
AU  - Giri P. Krishnan
AU  - Maxim Bazhenov
TI  - Can sleep protect memories from catastrophic forgetting?
AID  - 10.1101/569038
DP  - 2019 Jan 01
TA  - bioRxiv
PG  - 569038
4099  - http://biorxiv.org/content/early/2019/03/07/569038.short
4100  - http://biorxiv.org/content/early/2019/03/07/569038.full
AB  - Previously encoded memories can be damaged by encoding of new memories, especially when they are relevant to the new data and hence can be disrupted by new training – a phenomenon called “catastrophic forgetting”. Human and animal brains are capable of continual learning, allowing them to learn from past experience and to integrate newly acquired information with previously stored memories. A range of empirical data suggest important role of sleep in consolidation of recent memories and protection of the past knowledge from catastrophic forgetting. To explore potential mechanisms of how sleep can enable continual learning in neuronal networks, we developed a biophysically-realistic thalamocortical network model where we could train multiple memories with different degree of interference. We found that in a wake-like state of the model, training of a “new” memory that overlaps with previously stored “old” memory results in degradation of the old memory. Simulating NREM sleep state immediately after new learning led to replay of both old and new memories - this protected old memory from forgetting and ultimately enhanced both memories. The effect of sleep was similar to the interleaved training of the old and new memories. The study revealed that the network slow-wave oscillatory activity during simulated deep sleep leads to a complex reorganization of the synaptic connectivity matrix that maximizes separation between groups of synapses responsible for conflicting memories in the overlapping population of neurons. The study predicts that sleep may play a protective role against catastrophic forgetting and enables brain networks to undergo continual learning.Significance Continual learning, free of catastrophic forgetting, remains to be an unsolved problem in artificial neural networks. Biological networks have evolved a mechanism by which they can prevent interference and allow continual learning throughout the life of the organism. Building upon a range of data suggesting importance of sleep in memory and learning, here we test a hypothesis that deep sleep may play a role in continual learning and protecting memories from catastrophic forgetting. Our results revealed that complex patterns of synchronized oscillatory activity in the thalamocortical network during deep sleep reorganize synaptic connectivity to allow for consolidation of interfering memories and to enable continual learning.