Papex

Open problem · Neuroscience · Open

Function of sleep

Why do animals need to sleep?

Summary

The biological function of sleep is unresolved; proposals include memory consolidation and metabolic clearance.

Why it is open

Sleep is a reversible state of reduced responsiveness, usually with immobility [1]. It looks costly: Jerome Siegel notes that it is often seen as "a vulnerable state that is incompatible with behaviours that nourish and propagate species" [2]. Yet Chiara Cirelli and Giulio Tononi, reviewing the species studied, mostly mammals and birds, found "no clear evidence of a species that does not sleep" [1], and a sleep-like state has been described in the jellyfish Cassiopea, which has a nerve net but no centralised nervous system [10]. Prolonged sleep deprivation kills rats and flies, but it is unclear whether death comes from the loss of sleep itself or from forced arousals and stress, and pigeons appear to survive it [1].

Several candidate functions have evidence: consolidating memories, clearing waste from the brain, renormalising synapses and saving or reallocating energy [4, 5, 6, 8]. Yet Cirelli and Tononi wrote in 2008 that "No current hypothesis can explain why animals need to sleep" [1], and Siegel that of the many theories, "none is well established" [2]. Some researchers doubt that there is a single vital function to find [1, 2]. Siegel proposes that sleep is a variant of the dormant states seen across plants and animals, adaptive because it "optimizes the timing and duration of behaviour", with ecological variables the main determinants of how much animals sleep [2].

No body sets a bar for a resolution. Cirelli and Tononi state one view of it: a core function, if it exists and serves the brain, should be one the brain "cannot fulfill during wakefulness", and "the burden of proof rests with those" who claim one [1]. Many of the experiments below were done in mice or rats; the species is given for each.

What is known

Sleep across species (2008–2017). Flies and zebrafish meet behavioural criteria for sleep, including more sleep after being kept awake [1]. When animals are kept awake for 6–24 hours, some sleep, usually 5–10% of the normal amount, always intrudes [1]. Dolphins show the slow waves of sleep in one brain hemisphere at a time; Cirelli and Tononi read this as evidence that sleep "cannot be eliminated", although whether it is sleep has been questioned [1]. In all species studied (flies, mice, rats, hamsters and sparrows), waking raises the activity of genes for energy metabolism, cellular stress and synaptic strengthening, while sleep raises genes for synaptic weakening and for building membranes [1]. Cassiopea jellyfish are quiescent at night, respond more slowly to stimuli while quiescent, and are less active and responsive the day after being kept from it [10].

Energy (2009–2014). Siegel cites a reduction of about 30% in the brain's energy use during non-rapid-eye-movement (NREM) sleep compared with quiet waking, and notes that in humans the brain is 2% of body weight but uses about 20% of the energy spent in quiet waking [2]. In seven healthy adults living in a whole-room calorimeter, energy expenditure over 24 hours rose by about 7% in the first day of sleep deprivation and fell by about 5% during recovery [3]. Markus Schmidt's energy allocation model proposes that sleep shifts energy from waking activity to processes that run during sleep, rather than only saving it [8].

Memory (1994–2013). Björn Rasch and Jan Born review evidence that sleep helps retain memories and that memories are actively reorganised during slow-wave sleep [4]. In a 1994 rat study by Matthew Wilson and Bruce McNaughton, hippocampal "place cells" that fired together while running on a track fired together again during the following slow-wave sleep, but not during sleep before the run; and in two later rat studies, suppressing hippocampal ripples (brief bursts of fast oscillation) after learning, without disturbing sleep, impaired spatial memory [4]. The idea that rapid-eye-movement (REM) sleep contributes to memory consolidation was prompted in particular by its wake-like brain activity and by "frequent reports of vivid dreams after awakening from REM sleep" [4]; whether dreaming itself has a function is a separate Papex problem, the function of dreaming.

Synapses (2014–2017). Tononi and Cirelli's synaptic homeostasis hypothesis (SHY) holds that learning during waking strengthens synapses, which raises energy needs and saturates learning, and that during sleep "spontaneous activity renormalizes net synaptic strength" [6]. With three-dimensional electron microscopy of 6,920 synapses in mouse cortex, Luisa de Vivo and colleagues found the contact area between axon and spine about 18% smaller after sleep than after waking, in proportion to synapse size, sparing large synapses that lacked recycling endosomes [9]. In mouse motor cortex, Guang Yang and colleagues found that sleep after learning promotes the growth of new dendritic spines on particular branches, and that disrupting the reactivation of learning-active neurons during NREM sleep prevents it [7].

Clearance (2013–2025). In mice, Lulu Xie and colleagues reported that natural sleep or anaesthesia increased the space between brain cells by 60%, increasing the exchange of cerebrospinal fluid with the fluid around cells and speeding the clearance of β-amyloid [5]; the proposed pathway is called the glymphatic system [12]. In people given a tracer into the cerebrospinal fluid for MRI, one night without sleep (7 people, against 17 who slept) impaired its clearance from most brain regions, and the next night's sleep did not make up for it [11]. In mice, Natalie Hauglund and colleagues found synchronised oscillations of noradrenaline (norepinephrine), blood volume and cerebrospinal fluid to be the strongest predictors of glymphatic clearance during NREM sleep, with rhythmic changes in blood vessel diameter pumping cerebrospinal fluid into the brain [13].

A contested result (2024–2025). Andawei Miao and colleagues injected a dye deep in the brains of male mice and measured it in the frontal cortex: its spread did not change between sleep, waking and anaesthesia, and clearance was "markedly reduced, not increased" during sleep and anaesthesia [12]. Virginia Plá, Maiken Nedergaard and colleagues replied that this conclusion "is not supported by the data presented": the study measured movement of the dye within the brain, not its removal, gave each anaesthetic as one dose so that mice were mostly recovering, and measured sleep after sleep deprivation [14]. Nicholas Franks and William Wisden answered that the objections are "misunderstandings, misleading, or simply incorrect", and that most studies reporting faster clearance in sleep measure tracer entering the brain, which they regard as an unproven proxy for clearance [15].

Main approaches

Memory consolidation. Sleep, especially slow-wave sleep, replays and reorganises recent memories for long-term storage [4]. Strongest evidence: replay of waking activity during sleep, and memory loss when ripples are suppressed after learning, both in rats [4]. Main difficulty: the causal evidence centres on hippocampus-dependent memories in rodents [4], and Siegel argues that proposed functions such as these, though they occur in sleep, "cannot explain the variation of sleep amounts" within and between species [2].

Synaptic homeostasis. Sleep is "the price the brain pays for plasticity": it renormalises synapses strengthened during waking [6]. Strongest evidence: smaller synapses after sleep in mouse cortex [9] and sleep-related gene expression across species [1]. Main difficulty: sleep after learning also builds new spines in mice [7], and Rasch and Born treat downscaling as a process that complements active consolidation, not the whole function [4].

Waste clearance. Sleep increases the flow of cerebrospinal fluid through the brain and removes metabolites [5]. Strongest evidence: faster clearance during sleep in mice [5], a pumping mechanism tied to NREM sleep in mice [13] and slower clearance after sleep loss in a small human study [11]. Main difficulty: another mouse study found the opposite [12], and the groups disagree about whether clearance should be measured as removal from the brain or as tracer inflow [14, 15].

Energy and adaptive inactivity. Sleep saves or reallocates energy and keeps animals inactive when activity is least useful [2, 8]. Strongest evidence: lower brain energy use in NREM sleep [2] and higher energy expenditure without sleep in people [3]. Main difficulty: the human measurements come from seven adults [3], and Cirelli and Tononi argue that the lack of non-sleeping species, the intrusion of sleep into enforced waking and one-hemisphere sleep in dolphins tell against sleep being merely useful inactivity [1].

Open sub-questions

  1. Two definitions, one experiment. In the same mice, do tracer removal from the brain and tracer inflow from cerebrospinal fluid change in the same direction between natural sleep, waking and anaesthesia [5, 12, 14, 15]? See also How general anaesthesia works.
  2. Human clearance. Does the effect of one night without sleep on tracer clearance [11] replicate in larger groups, and does it scale with the NREM sleep lost [13]?
  3. Shrinking and growing synapses. In the same cortical area and sleep period, how do the general shrinkage of synapses [9] and new spines on learning-activated branches [7] coexist?
  4. Beyond the hippocampus. Does disrupting sleep replay impair memories that do not depend on the hippocampus [4]?
  5. A sleeper without a brain. Does depriving Cassiopea of night-time quiescence change synaptic or metabolic markers in its nerve net, testing whether synaptic [6] or energy [8] accounts extend to an animal without a centralised nervous system [10]?
  6. REM and dreaming. Does REM sleep have a memory role separate from slow-wave sleep, and does dream content track it [4]? See Function of dreaming.

Work on this in Papex

If you have a testable answer to one of the sub-questions above, you can draft it as a private hypothesis in Papex, or prepare the draft from Claude or ChatGPT with the Papex connector. You review the exact version, sign it and decide whether to publish it; a public record that names this problem is listed on this page. A Papex record shows who signed which version and when. It is not peer review and does not mean the hypothesis is correct.

Key references

  1. Cirelli, C. & Tononi, G. (2008). Is sleep essential? PLoS Biology 6(8), e216. 10.1371/journal.pbio.0060216 · Checked 14 Oct 2026
  2. Siegel, J. M. (2009). Sleep viewed as a state of adaptive inactivity. Nature Reviews Neuroscience 10(10), 747–753. 10.1038/nrn2697 · Checked 14 Oct 2026
  3. Jung, C. M., Melanson, E. L., Frydendall, E. J., Perreault, L., Eckel, R. H. & Wright, K. P. (2011). Energy expenditure during sleep, sleep deprivation and sleep following sleep deprivation in adult humans. The Journal of Physiology 589(1), 235–244 (published online 2010). 10.1113/jphysiol.2010.197517 · Checked 14 Oct 2026
  4. Rasch, B. & Born, J. (2013). About sleep's role in memory. Physiological Reviews 93(2), 681–766. 10.1152/physrev.00032.2012 · Checked 14 Oct 2026
  5. Xie, L., Kang, H., Xu, Q., Chen, M. J., Liao, Y., Thiyagarajan, M. et al. (2013). Sleep drives metabolite clearance from the adult brain. Science 342(6156), 373–377. 10.1126/science.1241224 · Checked 14 Oct 2026
  6. Tononi, G. & Cirelli, C. (2014). Sleep and the price of plasticity: from synaptic and cellular homeostasis to memory consolidation and integration. Neuron 81(1), 12–34. 10.1016/j.neuron.2013.12.025 · Checked 14 Oct 2026
  7. Yang, G., Lai, C. S. W., Cichon, J., Ma, L., Li, W. & Gan, W.-B. (2014). Sleep promotes branch-specific formation of dendritic spines after learning. Science 344(6188), 1173–1178. 10.1126/science.1249098 · Checked 14 Oct 2026
  8. Schmidt, M. H. (2014). The energy allocation function of sleep: a unifying theory of sleep, torpor, and continuous wakefulness. Neuroscience & Biobehavioral Reviews 47, 122–153. 10.1016/j.neubiorev.2014.08.001 · Checked 14 Oct 2026
  9. 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), 507–510. 10.1126/science.aah5982 · Checked 14 Oct 2026
  10. Nath, R. D., Bedbrook, C. N., Abrams, M. J., Basinger, T., Bois, J. S., Prober, D. A. et al. (2017). The jellyfish Cassiopea exhibits a sleep-like state. Current Biology 27(19), 2984–2990. 10.1016/j.cub.2017.08.014 · Checked 14 Oct 2026
  11. Eide, P. K., Vinje, V., Pripp, A. H., Mardal, K.-A. & Ringstad, G. (2021). Sleep deprivation impairs molecular clearance from the human brain. Brain 144(3), 863–874. 10.1093/brain/awaa443 · Checked 14 Oct 2026
  12. Miao, A., Luo, T., Hsieh, B., Edge, C. J., Gridley, M., Wong, R. T. C., Constandinou, T. G., Wisden, W. & Franks, N. P. (2024). Brain clearance is reduced during sleep and anesthesia. Nature Neuroscience 27(6), 1046–1050; author correction 27(7), 1425 (source data for Fig. 3b). 10.1038/s41593-024-01638-y · Checked 14 Oct 2026
  13. Hauglund, N. L., Andersen, M., Tokarska, K., Radovanovic, T., Kjaerby, C., Sørensen, F. L. et al. (2025). Norepinephrine-mediated slow vasomotion drives glymphatic clearance during sleep. Cell 188(3), 606–622. 10.1016/j.cell.2024.11.027 · Checked 14 Oct 2026
  14. Plá, V., Kroesbergen, E., Deng, S., Giannetto, M. J., Hablitz, L. M., Newbold, E. et al. (2025). A curious concept of CNS clearance. Nature Neuroscience 28(4), 731–733 (Matters Arising). 10.1038/s41593-025-01897-3 · Checked 14 Oct 2026
  15. Franks, N. P. & Wisden, W. (2025). Reply to: A curious concept of CNS clearance. Nature Neuroscience 28(4), 734–736. 10.1038/s41593-025-01898-2 · Checked 14 Oct 2026

Page reviewed 14 Oct 2026

Source

Open, as listed by Wikipedia · Unsolved problems in neuroscience. Checked 29 Sep 2026.

The status is the source’s. Papex does not decide whether a problem is solved.

Records on this problem

No public record addresses this problem yet.

More in Life sciences