Despite the fact that sleep is essential to our health, its function and what makes it necessary have remained mysterious. Over the years, scientists have accumulated data showing that sleep, or the lack thereof, affects the brain. Most of this work focused on the idea that sleep is important for consolidating newly formed memories. However, evidence is now building that sleep also makes room for the formation of new memories, acting as a sort of “spring cleaning” for the brain. The idea that sleep may help balance brain resources and space is known as the homeostatic theory of sleep. While you are awake, your brain is constantly exposed to new information coming in from your senses. This information allows you to form new memories and learn about the world around you.  Interacting with the environment also causes changes in your brain – its cells branch and grow, and new connections are formed between them. These changes require energy as well as physical space, both of which are limited.

The homeostatic theory of sleep proposes that sleep prunes out unnecessary connections between cells, freeing up resources and space for new learning to take place the following day. This hypothesis leads to several important predictions. First, your sleep need should increase if you are in a more intense environment, where you are exposed to many new things. Second, cells in your brain should be larger and more complex before sleep than after it. Lastly, your ability to learn should be enhanced after a period of sleep. Three recent studies have provided evidence to support all of these predictions, greatly strengthening the homeostatic theory – and giving us yet another reason to hit the pillows.

Dr. Matthew P. Walker at the University of California, Berkley found that sleep can enhance fact-based memory capacity [1,2]. In this study, people were tested for their ability to memorize and then immediately recall 100 names and faces, once at noon and once at 6 pm. Half of the participants were allowed to nap between memorization sessions while their sleep was monitored using an electroencephalogram, which allows for the identification of different sleep states. The other half of the participants remained awake. For the second session, the learning ability of those in the nap group was about 18% higher than those in the awake group, with the performance of the nap group increasing and that of the awake group decreasing. In addition, those in the nap group who spent the most time in non-rapid eye movement (non-REM) sleep tended to show the highest learning improvement. Non-REM sleep is one of the two major stages of sleep, and is thought to be the most important stage for homeostatic sleep changes. While this study supports one of the predictions of the homeostatic theory – that the capacity to learn is enhanced after sleeping – it does not address what actually occurs in the brain during sleep to allow for enhanced learning. To address this question, we must change species, and examine the diminutive, yet perhaps more informative, fruit fly.

While fruit flies are often seen as nothing more than an annoyance, scientific research on the fly has yielded a number of important advances in our understanding of the brain. This may seem odd – after all, the brain of the fly has little in common with the brain of a human. However, the fly must still move, eat, see, feel, and yes, even sleep, and cells within their brains are actually very similar to the cells in ours. Of course, fruit fly brains have far fewer cells than human brains.  However, when you want to look at individual cells, this can be an advantage. In order to test whether sleep produces physical changes in individual cells, Dr. Chiara Cirelli, of the University of Wisconsin, measured the volume, width, length, and special features of brain cells from flies that were either asleep or awake []. After sleep, cells were about 20% narrower, with shorter branches. Interestingly, when flies were placed in a stimulating environment, with a number of different flies and space to explore, branch length increased by over 50%, with a similar increase in branch complexity. Branch complexity is a measure of the number of points at which a branch occurs. When these flies were then allowed to sleep, they slept more, and the more sleep a fly got, the lower the branch complexity became. This study provides compelling evidence for the homeostatic sleep theory, but does not address whether these changes are linked to learning ability. For that, we turn to a study from the lab of Dr. Paul Shaw, of Washington University in St. Louis.

Figure 1. Drawings representative of fly brain cells before and after sleep. The cell on the left shows the increased branch complexity and length found when the fly is exposed to an enriched environment. The cell on the right shows the smaller volume typical of cells after sleep. (Click to enlarge.)

The expression “you can bring a horse to water but you can’t make it drink” aptly describes one of the main frustrations of sleep research – without chemical intervention, which can cause a host of side effects (making data difficult to interpret), there is no way to induce natural sleep. Humans can be asked nicely to go to sleep, and, given a dark quiet room, will sometimes oblige, but flies are a different story.  Fortunately, Dr. Shaw was able to identify a group of cells in the fly brain which, when activated, put the flies to sleep. To test the homeostatic sleep theory, Shaw tested how well a fly is able to learn a new task after high environmental stimulation, both before and after sleep [4,5]. As in Cirelli’s experiments, flies were placed in an enriched social environment, which led to an increase in the number of connections between cells. When flies were then trained in a memory task after being housed in this environment, they were not able to form lasting memories. Yet, if sleep was induced between exposure to a stimulating environment and the memory task, learning was restored. Additionally, when the cells from these flies were examined, the levels of cellular connections had returned to normal after sleep.

Taken together, these studies suggest that sleep is necessary for the brain to perform at its best. This seems to occur because the brain is a physical entity, which needs to be cleaned up and organized for new information to be properly absorbed. This mental spring cleaning process involves paring down the number of connections between cells, to keep the necessary and get rid of the unimportant. This pruning translates into a renewed ability to learn the next day. Sleep, far from being an indulgence, is essential for our brain to learn and remember. I, for one, am glad for the excuse.

Rebecca Reh is a Ph.D. candidate in the Program in Neuroscience at Harvard Medical School.


1. Bryce Mander, Sangeetha Santhanam, Jared Saletin and Matthew Walker. Wake deterioration and sleep restoration of human learning. Current Biology 2011.

2. Robert Preidt ‘Brain’s Learning Ability Seems to Recharge During Light Slumber’.

3. Daniel Bushey, Guilio Tononi, and Chiara Cirelli. Sleep and Synaptic Homeostasis: Structural Evidence in Drosophila. Science June 24, 2011.

4. Jeffrey Donlea, Matthew Thimgan, Yakuko Suzuki, Laura Gottschalk and Paul Shaw. Inducing Sleep by Remote Control Facilitates Memory Consolidation in Drosophila. Science June 24, 2011.

5. Washington University School of Medicine. “Sleep switch found in fruit flies.” ScienceDaily, 23 Jun. 2011.

Leave a Reply

Your email address will not be published. Required fields are marked *