Verified by 
![//creativecommons.org/licenses/by-sa/4.0)], via Wikimedia Commons](https://cdn.psychologytoday.com/sites/default/files/styles/article-inline-half/public/field_blog_entry_images/synapse%20image%20032617_0.jpg?itok=1bXm3kcV "//creativecommons.org/licenses/by-sa/4.0)], via Wikimedia Commons")
We continue to debate the purpose and need for sleep despite over a hundred years of scientific study. The first book to approach sleep from a physiological viewpoint was “Le probleme physiologique du sommeil,” by the French scientist Henri Pieron (Howell, 1913). The beginning of sleep medicine can be traced to Loomis, who, in 1937 documented the EEG characteristics (such as sleep spindles and K complexes) of what is now called non-REM sleep (Shepard, et al., 2005).
Many theories have been put forth as to why we would need a period of daily sleep that means being vulnerable and unconscious for a significant part of the day. Every animal studied thus far appears to need sleep. Even fruit flies (Drosophila melanogaster) show the characteristics we associate with sleep in vertebrates and mammals. As examples, they have sustained periods of quiescence with an elevated threshold to arousal - especially at night, are affected by the same stimulants and hypnotics that work for humans, and show great variability between individuals in the amount and timing of sleep, just as in humans (Cirelli & Bushey, 2008).
Some of the most common reasons proposed for the need for sleep are energy conservation, hiding from predators, and allowing the brain to clean up after a day of intense cognitive activity. These theories focus on either the restorative or the information processing aspects of sleep. In the restorative category, sleep is thought to be involved in the repair of cellar structures, the replenishment of needed chemicals, and the removal of waste products (Acsády, & Harris, 2017). For example, Xie et al. (2013), showed that neural metabolic waste products were cleared out of the brain at a much higher rate during sleep than during the awake state. Information processing theories suggest that sleep functions to support computational processes such as the consolidation of memory (Acsády, & Harris, 2017). Accordingly, sleep has been linked to the underlying processes needed to carry out various aspects of cognition, including forgetting unnecessary information accumulated during the day. For example, Crick and Mitchinson (1983) proposed that one of the functions of REM sleep may be to remove unhelpful information that accumulates during the day. The vivid images and strange story lines of dreams might be related to this review of and clearance of information. They termed this process “reverse learning”.
More recently, attention has been focused on the non-REM aspects of sleep. For example, sleep spindles, which occur predominantly in non-REM sleep stage 2, have been linked to our ability to transfer information within areas of the cortex and so to learn. We know that sleep deprivation results in significant disruption of cognition (Acsády, & Harris, 2017), as anyone who has pulled an “all-nighter” knows.
We do not know if any of the existing theories regarding the need for sleep are correct. Sleep has probably evolved to serve many physical and mental functions. Several recent studies have shown that at both the structural and molecular level, synapses undergo significant change during sleep.
Synapses are the gaps between nerve cells that allow for communication between the cells. When one nerve cell sends a message to another cell it is usually accomplished by the release of neurotransmitters that affect the electrical properties of the downstream cell. The axon is the part of the nerve cell that carries information away from the cell body to the synapse between it and another cell’s dendrites. Dendrites are the branches of a nerve cell where information from other nerve cells is collected across the synapse. Spines on dendrites facilitate this collection. Nerve cells are organized into networks that allow for complex information processing and learning. Keeping this system provided with necessary chemical and energy supplies as well as clearing away the metabolic wastes of this processing may be facilitated by the brain being in a relatively non-responsive state as during sleep. Alterations in the structure of synapses may directly impact on the processes of learning and memory.
A study by de Vivo et al. (2017) used scanning electron microscopy on the somatosensory and motor areas of the mouse brain to study changes in dendritic spines. It was observed that during sleep the dendritic spines decreased by about 18% when compared to their size during wakefulness. This shrinkage is probably associated with a decrease in synaptic strength. The downscaling of synapses was not, however, uniform. Large synapses were not affected, while smaller ones were. This could indicate that important memory traces, deeply encoded in the larger and more developed synapses, are not decreased during sleep, while weaker and less important information encoded in the smaller synapses is selectively removed.
Spine size and glutamate receptors were also found to be decreased during sleep in a study by Diering et al. (2017). The weakening of the synapses occurred in part through the active removal of a specific type of glutamate receptor. (Glutamate receptors are excitatory in nature.) They found that the neurotransmitters noradrenaline and adenosine were the drivers of these changes, which were mediated by a gene known as Homer1a. The concentration levels of these two neurotransmitters alternate during the sleep/wake cycle. (Adenosine is involved in drowsiness and noradrenaline in wakefulness.) This resulted in strengthened synaptic connections during wakefulness and reduced connections during sleep. These findings are consistent with a model of wake and sleep in which synaptic connections are strengthened during the day as learning occurs and weaken at night during sleep when unnecessary information is removed.
This makes sense in that the brain could become overloaded with rigid and inflexible memories and thus become dysfunctional if less important information could not be cleared. The brain needs a mechanism to preserve the most important learned information while getting rid of unnecessary information. It would, for example, be deleterious if we couldn’t remember the words of our language or how to eat, and indeed, we would take this as a sign of brain dysfunction, as in dementia. So the preservation of the large synapses could be a basis of this necessarily resilient memory. We can also imagine the problem we might encounter if we could remember every single place in which we had parked our car over the years and had to select our most recent parking spot from decades of memories of car parking. If all our memories were of equal strength, this could be a herculean or even impossible task! Clearly, this would be counterproductive and weakening those unneeded memories is very helpful. It is thus very important that the brain has a way of selectively strengthening and also weakening synaptic connections so that memory can remain flexible and stable as well.
This research may have implications for understanding the effects of sleeping medications and may help inform the development of new ones. We certainly know from day to day experience that lack of sleep negatively affects thinking and learning and that good sleep helps. We also know that many patients using sleeping medication report difficulty with memory and lack of focus during the day.
So there is mounting evidence that sleep is critical to the biological machinery of learning and cognitive processing. Scientists are beginning to unravel the mystery of how this occurs at the cellular level, but much more remains to be discovered. One thing we do know for sure: Getting our “beauty sleep” has positive benefits beyond those we see in our bathroom mirror. We see them in the classroom, on the job, and while simply going about the many tasks of daily life. And this happens, at least in part, thanks to our incredible shrinking synapses!
Acsády, L. & Harris, K.D. (2017). Synaptic scaling in sleep. Science, 355 (6324), p. 457. doi: 10.1126/science.aam7917
Cirelli, C., & Bushey, D. (2008). Sleep and wakefulness in Drosophila melanogaster. Annals of the New York Academy of Sciences, 1129, p. 323–329. doi.org/10.1196/annals.1417.017
Crick, F. & Mitchinson, G. (1983). Nature, 304, p. 111 - 114 (14 July 1983); doi:10.1038/304111a0
de Vivo, L., Bellesi, M., Marshall, W., Bushong, E.A., Ellisman, M.H., Tononi, G. & Cirelli, C. (2017). Ultrastructural evidence for synaptic scaling across the wake/sleep cycle. Science, 355 (6324), p. 507-510. doi: 10.1126/science.aah5982
Diering, G.H., Nirujogi, R.S., Roth, R.H., Worley, P.F., Pandey, A. and Huganir, R.L. (2017). Homer1a drives homeostatic scaling-down of excitatory synapses during sleep. Science, 355 (6324), 511-515. doi: 10.1126/science.aai8355
Howell, W.H. (1913). Book review of Le problème physiologique du Sommeil, Science, 37 (953), pp. 525-526. doi: 10.1126/science.37.953.525
Shepard, J. W., et al. (2005). History of the Development of Sleep Medicine in the United States. Journal of Clinical Sleep Medicine, 1 (1), 61–82.
Xie, L., et al. (2013). Sleep drives metabolic clearance from the adult brain. Science, 342 (6156), pp. 373-377. doi: 10.1126/science.1241224
