How is Memory Consolidated During Deep Sleep?

Understanding how memory is recorded in the brain has been one of the most challenging problems in neuroscience and lead to implications for treating conditions with memory impairment. Sleep is an essential activity for memory consolidation for long-term storage, exactly how brain activity is recorded as long-term memory that can be recalled or replayed in the brain is previously not well understood and hard to decipher in experimental models. Researchers from the University of California at Riverside developed a mathematical model that offered insights into how this happens. During wakefulness, when the brain is receiving sensory input from one’s experiences, these episodic or short-term memory signals are created in a region of the brain called the hippocampus. Electrical activity stored in the hippocampus are progressively transferred to the outer layer of the brain called the cortex for storage as long-term memory, which are encoded by changes in the synaptic connections or wiring between brain cells in the cortex. This computational model revealed how short-term memory signals from the hippocampus replayed during deep sleep determine when and where the signal patterns in the cortex are formed. This in turn leads to permanent changes to the synaptic connections. These synaptic changes can reinforce the signal patterns in the cortex to create a replay of the same specific memory signals without further input signals from the hippocampus, demonstrating creation of long-term memory. This computational model provides a mechanistic explanation of how characteristic brain activity during deep sleep may promote consolidation of recent memories for long-term storage. In addition, it can further be used to make experimental predictions and develop interventions to address memory deficits.