The science of sleep

Birds do it, bees do it, it is entirely possible that fleas do it, and most of us would like a little bit more of it especially on a Monday morning. But what is it that makes a spot of shut-eye so vital to life?

We all know how important sleep is. Even if we don’t entirely know what it is for, most of us are all too familiar with the waves of tiredness that descend when we have stayed up a little too late, or even when we are faced with a less than exciting presentation. The nodding head and drooping eyelids inevitably turn our thoughts to a cosy bed and forty winks. What you might not know, is that with this propensity to sleep you are carrying on a millennia old tradition that has endured throughout the evolution of many species. Indeed, several sleep researchers believe that the need for sleep descends deep into the animal kingdom, suggesting that human sleep has a long and rich evolutionary history.

This makes the question of the function of sleep even more intriguing – any explanation now has to be universal enough to encompass the other slumbering species as well. This interspecies sleeping club doesn’t just include vertebrates either. In recent years scientists have found that invertebrates like honeybees and crayfish sleep as well, indeed the most extensive work has been carried out on fruit flies. It turns out that they rest for 10 hours a night, and if you keep them awake longer they need to sleep more.

So just what is sleep for? Well, two thousand years ago Aristotle thought that sleep resulted from warm vapours rising in the stomach. Perhaps if the trend for groups of men to eat their own weight in hot Indian food and continue to do so into the small hours was popular in pre-christian Greece, he may have disregarded this theory, but in fact the idea stuck for a while. In modern times however, the role of sleep has proved a tricky question to answer. “Sleep has attracted a tremendous amount of attention in science, but we really don’t know what sleep is,” said Steven Lima, a biologist at Indiana State University. “One of the reasons we don’t understand it is that we haven’t taken this evolutionary perspective on it.”

One thing that is clear is that it is not just the need to rest - the need to shut down - that drives the need for sleep. In fact, during the REM stage of sleep the brain is a hot bed of organised physiological activity, even if you are not aware of it. As Dr Allen Rechtschaffen, a sleep expert at the University of Chicago says: “You can rest all you like and you still need sleep.”

Perhaps the oldest theory put forward to explain the need to sleep is that it provides a period for the body to recover from the wear and tear of being awake. However, in reality researchers have yet to find any vital biological function that sleep restores. Jim Horne, Director of the Sleep Research Centre at Loughborough University explains: “All the evidence shows that organs outside the brain undergo their restoration equally effectively during relaxed wakefulness. During human night time sleep, blood amino acid levels usually fall owing to the absence of eating, and consequently, tissue repair is reduced.”

Another theory maintains that sleeping is just a way to protect animals by taking them out of harms way when predators roam. This idea, however, can’t explain why the sleep lost one night can be made up the next night or why the impact of long-term sleep deprivation is so severe – rats deprived of sleep actually die in two to three weeks.

Riding the wave: What your brain does when the lights go out

When most mammals sleep brain activity cycles between two distinct phases: REM (Rapid Eye Movement) and non-REM. In humans this cycle take around 90 minutes

Non - REM This phase is made up of 4 stages and accounts for 75 - 80% of sleep time.  - Stage 1. With the near disappearance of the Alpha waves seen during awake states, this stage is thought of as the transition to Stage 2. - Stage 2. The EMG lowers and conscious awareness of the external environment disappears. - Stage 3. Considered part of slow wave activity (SLA) and is primarily a transition period into Stage 4. - Stage 4. Often described as the deepest stage of sleep, consists of mainly SLA. It is difficult to wake someone up in this stage. It is this stage that sleep walking and sleep talking occur.

REM Predominant in the final third of sleep and associated with dreaming. Brain activity is similar to wakeful states and Stage 1.

The theories that have gained most attention in recent years all concern the formation and consolidation of memory in some way – how sleep does this however, is still very much up for debate.

Most mammals cycle between two distinct phases of sleep, one of which is characterised by rapid eye movement - REM sleep, and the other - non-REM - which is characterised by the lack of this movement. Humans tend to take about 90 minutes to complete a full cycle of REM and non-REM sleep.

For a long time sleep research focused on the REM phase, largely because this is where the most brain activity takes place. It was this phase, so the theories went, that was most important in memory consolidation and learning, largely because - unlike non-REM sleep - most neurons in the brain fire actively. It seemed to make sense that the phase of sleep where the brain was most active was the phase responsible for such a complex neuronal function as memory. Indeed, as technology developed and allowed more in-depth studies the idea seemed to hold.

Bruce McNaughton, a physiologist and psychologist from the University of Arizona and Texas, uses electrodes small enough to record the activity of single neurons. By recording the activity of up to 100 hundred separate neurons during the sleep wake cycle of a rat he thinks he may have found the role of REM sleep. His recordings show that many of the same neurons that fire when a rat is performing a task during the daytime are reactivated during REM when the rats sleep. This, he claims, is because the brain is reviewing its recently stored data so that it can be consolidated into permanent neuronal connections.

So far so good. However, there was a problem. People can be deprived of REM sleep for several months at a time, as occurs through the use of drugs like tricyclic antidepressants, yet show absolutely no memory impairment. Undoubtedly a blow to the REM memory consolidation theory. Similarly, patients with brain injuries that cause them to skip the REM phase of sleep appear to suffer no memory impairment. One Israeli man suffered a shrapnel injury leaving him completely REM deficient yet had a memory good enough to see him through law school.

So why would the brain bother with REM? This still seems to be a mystery. One thing that is certain is that dreams occur during REM. Dreams total around 100 minutes a night, but we can seldom recollect more than a few minutes worth. It is tempting to assume that dreams are the visual byproducts of memory consolidation - a kind of ‘cerebral dust’ that is kicked up when the neurons that have been activated during the day are switched back on to allow memory formation. However, as the knowledge of why we sleep builds, it seems that is a grossly over simplistic assumption. Jim Horne suggests: “Dreams are probably meant to be forgotten and have little more than entertainment value, simply being the nightly ‘cinema of the mind’.”

Whatever its function, REM sleep seems to be more prolific in the developing foetus. Horne thinks that the most plausible theory for this is that REM activity aids brain development by providing some sort of substitute stimulation in the relatively stimulation free uterus.

Despite the fact that REM-deficient people had perfectly fine memory function, the idea that sleep must play a role in memory consolidation still loomed large in the sleep research community. But if REM wasn’t responsible, what was?

Whilst studying the phase of non-REM sleep known as slow-wave activity (SLA) one researcher has come up with an explanation that was the complete opposite of what everyone expected. It suggests that rather than being a period of frantic neuronal circuit building, sleep based memory formation could actually rely on a ruthlessly efficient energy cost-cutting exercise.

Giulio Tononi, a neuroscientist at the University of Wisconsin, published his theory in Nature and thinks it could change the way we view SLA. During SLA, billions of neurons undergo a synchronous one second burst of electrical activity – the longer a person has been sleep deprived the bigger the initial burst. Tononi noticed that after each wave of synchronous activity the brain goes completely silent. This ‘on-then-off’ activity is a perfect way for the brain to weaken synapses – connections between neurons. With-out this weakening, thinks Tononi, the brain would face an energy crisis, needing more energy than could be supplied by the body.

During waking hours sensory experiences set vast numbers of neuronal circuits into action. This activity strengthens the synapses, and even forms new ones. “Normally the brain takes up 20% of the energy in the entire body. So by the end of the day, if you have synapses that are much stronger, the cost of running the brain is much higher,” explains Tononi. So to avoid a complete bottoming out of the energy supplies of the body, some of these synapses must be destroyed – they are simply too costly. With its synchronised firing, SLA appears to be the perfect tool to hunt out and streamline the neuronal connections, and may even allow the weakest ones to drop out.

The theory is now gaining weight, and many think that perhaps this represents the main function of sleep. By trimming the neural connections built up during the day, only the strongest synapses remain so that when we wake the weakest connections have vanished - and only the strongest circuits have been consolidated into memory. In many ways SLA acts as the doorman at the gate of memory – and only the strongest circuits are getting in.

Perhaps this also explains the function of REM. REM generally comes after period of intense SLA, could it be that REM is kick starting those circuits that are strong enough to survive the synaptic killing fields of SLA ensuring that they remain active and connected?

Tononi thinks that this basic principle of sloughing off weak synapses could be something that is common to all animals that sleep. “It would be a basic thing that would apply to any brain that can change.” he says.

So it seems that even though experts are closing in on the role that sleep has to play, there are many mysteries that still surround this age old practise. But given what they have learnt in the past few years one thing is certain – sleep on it, things will be so much clearer in the morning. 

Phil Prime.  Phil is assistant editor of Laboratory News. He holds a degree in Neuroscience from Nottingham University and an NCTJ in journalism from City College, Brighton.