Evidence Has Been Put Forth

¶ … evidence has been put forth by researchers in that there is a relationship between sleep, learning, and memory. The relationship was discussed by several German authors - Born, J., Rasch, B., Gais, S., in a review published in the Neuroscientist (2006). The review, called "Sleep to remember," starts with the assumption that sleep is a state that leads to (optimizes) the consolidation of newly acquired information in memory. The authors bring forth evidence that sleep mediates the storage of newly encoded information into long-term memory. There are two important constructs discussed in this article: memory and sleep characteristics related to memory.

The authors begin their review by defining the three fundamental processes concerning memory: acquisition, consolidation, and retrieval. Acquisition circumscribes the learning process (encoding process), while consolidation includes the processes that occur after learning is finished and leads to stabilization, and enhancement of the newly encoded information. Retrieval consists of recalling of the stored memories. The authors focus mainly on the consolidation process, which apart from the role of strengthening memory traces plays an important part in the integration of newly encoded pieces of information with preexisting long/term memories. Regarding the neural mechanisms underlying consolidation, there are two types of consolidation: "synaptic consolidation," taking place in the first minutes or hours after learning, involving the stabilization of synaptic change, and a "system consolidation," which requires days or months to complete, involving the reorganizations of memory representations in different neural networks. The reorganization during consolidation has been mostly reported for declarative memory, while for procedural, nondeclarative memory only hints have been made. The two memory systems rely on two separate neural systems: the declarative memory relies on hippocampus, while procedural memory relies on striatocortical circuitry.

After this short account on memory processes and systems, the researchers take into account the relationship between sleep and procedural and declarative memory. The type of task that mediates the relationship between the two constructs is taken into consideration. Studies have proved that for tasks in which visual texture discrimination is involved a gain in skill performance (the skill becomes stable and enhances) at later retrieval testing emerges only if the training is followed by a period of sleep (Gains and others 2000, Stickgold and others 2000, Mednick 2002, 2003, cited by Born et al.). Sleep dependent improvements were noticed in tasks involving finger sequence tapping (enhancement appears only after a period of sleep) (Fischer and others, 2002, Walker and others 2002, cited by Born et al.). Considering this evidence Walker suggested a difference should be made between two stages of consolidation in procedural memory formation: first, a stage of stabilization that takes some time and does not require sleep, and second, enhancement, that does require sleep. Taking into consideration the facts mentioned, the authors conclude that both processes - stabilization and enhancement - depending on the task characteristics are not exclusively determined by sleep but are clearly amplified by sleep. The sleep-specific reorganization of skill representations has been demonstrated using functional magnetic brain imaging.

Studies have also attested the presence of a positive relation between sleep and declarative memory. Tasks based on word recognition and word pair association indicated that sleep after learning improves consolidation in the declarative memory. As opposed to procedural memory which requires repeated trials but fades more slowly, declarative memory is faster and more sensitive to interference. Sleep has been proved to consolidate declarative memories by leading to a reorganization of the newly acquired representations, leading to a qualitative change in memories as well. Empirical evidence was put forth by using the task based on number reduction originally developed by Thurstone and Thurstone (1941) and serial reaction time task (SRTT) (Wagner and others 2004, cited by Born et al.). The idea underlying these types of task is that sleep favors the activation of an explicit knowledge of the sequence, while the training itself help participants gain only an implicit knowledge of the sequence.

There is another issue raised by Born et al. In their article, concerning the role of intention in the relation sleep - memory and concluded that supposedly the extent to which learning is intentional might eventually determine whether memories gain from sleep or not.

The sleep stages play also an important part in memory consolidation. The debate in research is whether memories are consolidated mainly in slow-wave sleep (SWS) (sleep stages 3 and 4) or REM sleep. However, the hypotheses which the authors adopt, supported by the results of retention tests, is that different types of declarative memories benefit greater from early SWS-rich periods of sleep, while procedural memories benefit from periods of late sleep (REM). Moreover, researchers acknowledge that recently acquired memories are reactivated during sleep, fact demonstrated in rats using hippocampus dependent tasks such as maze learning. Reactivation was observed mainly during SWS, and in some cases in REM as well. The inference that the authors draw from here, along with other researchers (Buzsaki, 1998, Gais and Born 2004 cited by Born et at) is that the process of consolidation for hippocampus-dependent memories can be conceptualized as an interaction between neocortical and hippocampal networks. Accordingly, new material is temporarily stored in hippocampus while the permanent transfer of information is being made in neocortical regions.

The neurotransmitting mediation that regulates the formation of hippocampus dependent memories was tested and confronted with previous findings. Acetylcholine (Ach) is supposed to regulate the interaction between neocortex and hippocampus. The researchers tested his model by infusing a choliesteraze inhibitor in participants - who on a night before learned a paired associate word lists - during a SWS nocturnal sleep. The enhancement in cholinergic activity was therefore enhanced, the fact leading to impaired recall. Blocking cholinergic activity in participants while awake improved retention, but impaired acquisition of new knowledge. These findings suggest that ACh is a regulator of the neocortico-hippocampal dialogue by switching between modes of acquisition and consolidation. In the same way, the influence of the hormone cortisol is shown to impact the dialogue above, enhancing the acquisition of new information while interfering with retrieval of old memories during wakefulness.

The last part of the study concerns the effect of slow oscillations on memory consolidation. Born et at hypothesize starting from previous research (Contreras and Steriade 1995, Meier-Koll and others 1999, Fogel and others 2002, Gais and others 2002, Clemens and others 2005, 2006) that slow oscillations have the function to promote a coordinated reactivation of memory representations and bind together different neurocortical networks - thalamic, hippocampal, brainstem structures like locus coeruleus.

The study ends up abruptly, after describing the way slow oscillations coordinate the reactivation of memory representations, without a unifying glimpse over the issue debated. The review is organized into several sections referring to: memory systems and processes, sleep-dependent consolidation of procedural and declarative memory, the reorganization of memory representations in sleep, the sleep stages involved in memory consolidation, neural structures involved in memory consolidation by reactivation of representations, the neurotransmitters and hormones that mediate memory consolidation and the role of coordination played by slow oscillations in memory consolidation. Therefore, the key aspects discussed in the article regard memory consolidation and sleep characteristics, with a focus on the first concept.

The overall impression with respect to the review "Sleep to Remember" is that it represents a combination of meta-analysis and individual research. The review of previous studies is interspersed with their own research, suggesting hypotheses and then trying to test them. However, a limitation of this study is that much of the research mentioned for which the Born et at claim authorship is done on rats, leaving a possibility for human replication (when it is possible and ethical). Moreover, the authors tend to consider only the arguments that support their hypotheses and overlook those which contradict them. A more comprehensive account on the mechanisms involved in the role of sleep in declarative memory consolidation is given by Born and Gais in an article from the year 2004. The main findings, are related to the reactivation of newly acquired memory representations in hippocampal networks that stimulates a transfer and integration of these representations into neocortical neuronal networks as the most important mechanism of memory consolidation. However, the authors give an account of these findings in the article discussed as well.