Sleep Microstructure and Memory Function

NoN-REM SlEEp, REM SlEEp, aNd MEMoRy Several studies investigated the effect of sleep on memory. In humans, arguably the first experimental description of a beneficial role of sleep for memory stabilization was provided in 1924 (1), indicating a protective benefit of sleep in preventing the normal decay-curve of forgetting that develops across time spent awake. Specific stages of sleep appear to be critical for memory consolidation. Memory is commonly divided into a declarative and a non-declarative memory system (2). Declarative memory is defined by memories accessible to conscious recollection, i.e., memories for events in a spatio-temporal context (episodic memory) and factbased information (semantic memory). Procedural memory for skills is the type of non-declarative memory most thoroughly studied with regard to the effects of sleep. Sleep is characterized by the cyclic occurrence of REM and NREM sleep comprising SWS (sleep stages 3 and 4) and lighter sleep stages 1 and 2. Two hypotheses were proposed regarding sleep stages and memory consolidation (3). The dual process theory assumes that the specific sleep stages support consolidation of different types of memories. SWS supports declarative memory consolidation whereas REM sleep does so for procedural memories (4). The sequential hypothesis, on the other hand, proposes that sleep benefits memory optimally through the cyclic succession of both SWS and REM sleep. The original version of this hypothesis assumed that SWS functions to weaken non-adaptive memory traces whereas REM sleep re-stores the remaining traces (5). The dual process hypothesis received support mainly based on the early late sleep comparison, i.e., an approach comparing effects of retention intervals covering the first (SWS-rich) or the second (REM sleep-rich) half of nocturnal sleep. SWS-rich early sleep consistently found to support consolidation of hippocampus-dependent declarative memories, whereas REM sleep benefited non-declarative types of memory like priming, and memories for visuo-motor skills (6, 7). However, this dichotomy does not fit all results. Several non-declarative tasks, like visual texture discrimination, are also supported by SWS whereas REM sleep in some instances seems to benefit aspects of declarative memory (8). More recently, Genzel and colleagues (9) tried to clarify the link between specific sleep stages and different types of memory consolidation by suppressing sleep stages. They deprived subjects once each of REM sleep and SWS, and once let them sleep undisturbed through the night. After each night, the authors tested declarative and procedural memory consolidation. Although REM sleep and SWS awakenings led to a significant reduction of the respective sleep stages, memory consolidation remained unaffected. According to the authors, there are two possible explanations for REM deprivation and SWS deprivation not influencing sleepdependent consolidation of motor tasks: (1) the diminished amount of REM sleep in the REM deprivation condition was still sufficient for sleep-dependent memory consolidation or (2) the memory consolidation is dependent on stage 2 sleep. Another explanation of the results is that sleep-associated processes contributing to memory consolidation requires analyses of polysomnographic phenomena, like sleep spindles and slow-wave activity (SWA). This aspect is very important also for the implications for the cognitive decline by increasing age (10, 11). Other factors that change with age, such as hormones and neurotransmitters, or hypoxia related to sleep-disordered breathing (12, 13), may also play an important role. SlEEp SpiNdlES aNd MEMoRy Sleep spindles – a hallmark of stage 2 – are characterized in humans by waxing and waning 11to 15-Hz oscillations lasting 0.5–3 s. They are generated by the thalamus, which acts as a pacemaker (14) and result from reciprocal rhythmic interactions between reticular and thalamo-cortical cells. Several reports described an association between NREM sleep spindle events and memory, including hippocampal-dependent learning. These short synchronous bursts of activity coincide with hippocampal sharp waves and ripples (15, 16), possibly reflecting reactivation of learned memory representations (17). Moreover, spindle density was associated with next-day memory recall: Meier-Koll and colleagues (18) reported a similar increase in spindles following learning of a hippocampally dependent maze task, and Clemens and colleagues (19) identified a correlation between spindle density and overnight verbal memory retention. The spindle-related memory processing is supported by very recent data showing that the same cortical areas active during learning are also (re)activated during post-encoding sleep spindle events (20). Sleep spindles can be separated into two subtypes based on frequency (21): “slow” (11–13 Hz) and “fast” (13–15 Hz). Concerning the memory consolidation, recent neuroimaging data demonstrated that the occurrence of fast sleep spindles coincide with moments of increased functional connectivity between the hippocampus and areas of neocortex (22), supporting a postulated component of the hippocampal-neocortical model of sleep-dependent memory consolidation. Indeed, several studies described select associations between episodic memory and fastbut not slow-spindle activity (23, 24). It is known that healthy aging is characterized by a reduced numbers of sleep spindles (25), and several studies reported reduced sleep spindles density in patients with Parkinson’s disease (26, 27).