|Dreaming: the Conscious Memory Consolidation||Memory|
Dreams occur mainly during the rapid-eye movement (REM) sleep and the non-REM (NREM) stage 2 sleep. They arise from the replay of pre-sleep experiences (Wamsley and Stickgold, 2011). The underlying mechanism could be the same as the spontaneous synaptic reactivation which is essential for memory consolidation.
Why We Are Conscious in REM and NREM Stage 2 Sleep
Dreams indicate that we are still aware of the mental processes while memory is being consolidated. We cannot experience any dreams during the slow wave sleep (SWS) (i.e., NREM stage 3 sleep) because SWS consolidates memory in the neocortex where the slow oscillations can cause unconsciousness (see Chapter 11). Evidence suggests that the amygdala memory is consolidated during the REM sleep (Menz et al., 2013), which is the lightest sleep. The amygdala is located deep in the brain. Its activities may have little impact on the integrity of the resting state networks.
Figure 12-1. The K-Complex. [Image source: Wikipedia]
The NREM sleep consolidates hippocampus-dependent memory (Marshall and Born, 2007; Wamsley et al., 2010). Its stage 2 is characterized by the K-complex recorded in electroencephalography (EEG). The K-complex represents a single cycle of the slow oscillation. It occurs sporadically during the NREM stage 2 sleep. These K-complexes could reduce the level of consciousness, but may not be sufficient to abolish it.
EEG reflects neuronal activities only in the cerebral cortex, not the deep brain areas such as the hippocampus and amygdala. The lack of a continuous slow wave during the NREM stage 2 sleep suggests that this stage may consolidate mainly the memory within the hippocampus, rather than the neocortex. Then, in the following stage 3, the hippocampus may send the consolidated signals to the neocortex.
Memory Consolidation in Hippocampus and Amygdala
Figure 12-2. Major steps in memory consolidation (see Chapter 10 for detail).
As in the neocortex, PKC also plays a key role for memory consolidation in the hippocampus and amygdala (Kim et al., 2013; Lai et al., 2008; de Quervain et al., 2012). Since reactivation is essential for consolidating memory in these regions, their underlying mechanisms could be basically the same. The consolidation processes are summarized in Figure 12-2. Reactivation in a brain area (neocortex, hippocampus or amygdala) will initiate consolidation in that area. The presynaptic signals may depend on the communication in the entire system.
After consolidation, PSD-95 will be able to translocate along microtubule tracks from the soma to target synapses and maintain their strength. More PSD-95 in the postsynaptic density (PSD) will retain more AMPAR at the postsynaptic membrane, thereby increasing the synaptic strength. On the other hand, PSD-95 can also recruit CaMKII which may further strengthen the synapse in response to stimulation. Moreover, tubulins can be recruited by CaMKII to inhibit synaptic activation. The levels of these molecules in PSD determine the ability to recall an event at a given moment, whereas the memory per se is encoded in the microtubule tracks transporting PSD-95.
Thus, the inability to recall an event does not mean that the memory has been lost. It could be due to the inhibition of tubulins or low level of PSD-95. The next chapter will show how the exodus of tubulins from PSD may cause us to remember past experiences when we die.
Author: Frank Lee