Quote:
EARLIER this year, a puzzling report appeared in the journal Sleep Medicine. It described two Italian people who never truly slept. They might lie down and close their eyes, but read-outs of brain activity showed none of the normal patterns associated with sleep. Their behaviour was pretty odd, too. Though largely unaware of their surroundings during these rest periods, they would walk around, yell, tremble violently and their hearts would race. The remainder of the time they were conscious and aware but prone to powerful, dream-like hallucinations.
Both had been diagnosed with a neurodegenerative disorder called multiple system atrophy. According to the report's authors, Roberto Vetrugno and colleagues from the University of Bologna, Italy, the disease had damaged the pair's brains to such an extent that they had entered status dissociatus, a kind of twilight zone in which the boundaries between sleep and wakefulness completely break down (Sleep Medicine, vol 10, p 247).
That this can happen contradicts the way we usually think about sleep, but it came as no surprise to Mark Mahowald, medical director of the Minnesota Regional Sleep Disorders Center in Minneapolis, who has long contested the dogma that sleep and wakefulness are discrete and distinct states. "There is now overwhelming evidence that the primary states of being are not mutually exclusive," he says. The blurring of sleep and wakefulness is very clear in status dissociatus, but he believes it can happen to us all. If he is right, we will have to rethink our understanding of what sleep is and what it is for. Maybe wakefulness is not the all-or-nothing phenomenon we thought it was either.
Received wisdom has it that at any given time, healthy people are in one of three states of vigilance: awake, in rapid eye movement (REM) sleep or in non-REM (NREM) sleep. Each state is distinct and can be recognised by a characteristic pattern of brain activity, as measured by an electroencephalogram (EEG) (see chart, right). Wakefulness is easy to detect. Apart from the fact that a person's eyes are open and they are responsive, their EEG shows a pattern of high-frequency, low-amplitude waves. NREM sleep is divided into four stages, each of which has its own distinctive EEG pattern. REM is trickier to spot because in EEG terms it closely resembles stage 1 NREM sleep. So to be sure it really is REM, researchers also look for the telltale rapid eye movements and a slackening in the muscles of the chin and jaw.
Mahowald is not the only person to have questioned these neat distinctions. David Dinges, a psychiatrist at the University of Pennsylvania, Philadelphia, has probably deprived more people of sleep in the name of science than anyone else. In one such study in the late 1980s, Dinges and his team revealed how easily the different states of vigilance can become intermingled. When volunteers were subjected to tests of working memory in which they had to subtract numbers, they could do an average of 90 sums in 3 minutes with few errors. After 52 hours deprived of sleep, their performance fell to around 70 subtractions, with not many more errors. However, after they had slept for 2 hours the change was dramatic. "When we woke them up abruptly, and they rated themselves as alert and ready to go, they couldn't do even one subtraction," says Dinges. People even seemed to be dreaming as they attempted the task. One subject mused: "What if people ran faster than normal people run home," in the middle of a string of incorrect responses.
Known as sleep inertia, a less extreme version of such disorientation is now generally recognised as the cause of the grogginess some people get after their alarm clock goes off. It is as if they are socially awake but functionally asleep; as if the brain circuits underlying responsiveness are up and running, but those mediating working memory are still offline. Mahowald is convinced that it is just one of many disorders that can be explained as a breakdown in the boundaries between sleep and wakefulness. He lists a whole raft of such conditions in the same issue of Sleep Medicine as Vetrugno's description of people with status dissociatus (vol 10, p 159).
One is REM behavioural disorder (RBD), in which people in REM sleep act out their dreams because the temporary paralysis, or cataplexy, that normally accompanies this state is replaced by the full mobility of wakefulness. In sleep paralysis the opposite is true. Here, cataplexy intrudes into wakefulness, and a person wakes to find him or herself unable to move. It is estimated that up to 40 per cent of people have experienced this disturbing phenomenon. Also surprisingly common are hypnagogic hallucinations - sensory illusions that occur on the cusp of sleep when the dreaming component of REM intrudes into wakefulness. Mahowald's list also includes sleepwalking, night terrors and narcolepsy, which is an inherent instability in vigilance state boundaries characterised by rapid cycling between states and the tendency to fall asleep mid-sentence. Controversially, the list also includes near-death experiences and alien abductions. It is no coincidence, he says, that alien abductions almost always occur in the recumbent position, in the transition from wakefulness to sleep.
It is no coincidence that alien abductions tend to occur in the transition from wakefulness to sleep
Sleep is a highly complex phenomenon in which changes in sensory, muscular, hormonal and neural systems must be coordinated to create a particular state of vigilance. "When you think about it, there are billions of people in the world who are shifting between wakefulness, REM and non-REM multiple times every 24 hours," says Mahowald. "It's amazing that the switching is as accurate as it is." This accuracy suggests that there is a strong adaptive drive to be fully in one state or another. But he also thinks that state dissociation - the presence of more than one vigilance state concurrently - is more common than anyone previously suspected.
The boundaries between sleep and wakefulness are particularly blurred when we are sleep-deprived. Around a decade ago, Dinges realised that although his sleep-deprived volunteers seemed to be awake, they were in fact experiencing momentary lapses, or microsleeps. Since then, he has discovered that these fleeting naps last between half a second and 2 seconds, and become increasingly frequent the longer we are deprived of sleep, until finally we cannot recover and nod off. Dinges sees them as the outward sign of a tug-of-war between neural systems that are trying to initiate sleep, and others that are trying to maintain wakefulness.
This chimes with the ideas of James Krueger of Washington State University in Pullman, who has argued that the individual processing units in the brain - known as cortical columns - fall asleep independently when they become tired. In his view, shifts between wakefulness and sleep occur when enough columns are in one state or the other (Nature Reviews Neuroscience, vol 9, p 910). Krueger believes this mosaic pattern of sleep explains sleep inertia and sleepwalking.
Some people are more prone to microsleeps than others. In a 2007 study, Dinges and his colleagues showed that there are enormous differences in people's ability to resist the lure of sleep when tired. Among a group of healthy, non-sleep-deprived adults, the differences in alertness are small. "But as soon as we provoke the system with some sleep deprivation, those differences get larger and larger," he says (Journal of Sleep Research, vol 16, p 170).
This has implications for all of us, especially those in professions where staying awake is a life-or-death matter (see "Ground control to Major Tom"). "If you're doing 100 kilometres per hour on a highway and you have a lapse, your fingers go lax on the steering wheel and you drift off the road at a 4-degree angle," says Dinges. "Two seconds is all it takes to be completely out of your lane." Up to 20 per cent of traffic accidents are fatigue-related.
Brain imaging has recently revealed a mental back-up system in people who remain alert when sleep-deprived (Journal of Neuroscience, vol 29, p 7948). While other people have reduced brain activity when tired, sleep-resistant individuals manage to maintain their brain activity levels. More interestingly, they also recruit new areas to help compensate for having been awake for so long. These people were selected for the study because they had a gene variant found in around 40 per cent of people that is thought to be associated with resistance to sleep deprivation. It seems likely that such people are also less prone to state dissociation, although this has not yet been tested.
Another group who appear to be more vigilant than most are people who suffer from primary insomnia - insomnia not associated with any other condition. There is evidence to suggest they are in a constant state of hyperarousal, with relatively high metabolic rates and high levels of the stress hormone cortisol. "It's as though somebody just cranked up their alertness over 24 hours, so they are more alert at night, but they are also more alert in the day," says Mahowald.
As the blur between sleep and wakefulness becomes more widely accepted, researchers are devising techniques for capturing the brain's fleeting lapses and vacillations. For example, neuroscientist Giulio Tononi of the University of Wisconsin, Madison, is eavesdropping on sleeping brains using EEGs with 256 electrodes rather than the more usual 32, to improve spatial resolution and help him catch the brain in the act of mosaic napping. Microsleeps are just the tip of the iceberg, Tononi says. He is particularly concerned by the possibility that parts of our brain might be going offline without us even realising it. "In many respects, it would be like having a temporary mental disorder without anybody, including yourself, being aware of it," he says. Forgetfulness and daydreaming could be examples of this, but so could more bizarre and even criminal behaviours (see "When a crime is not a crime").
It is like having a temporary mental disorder without anyone, including yourself, being aware of it
Meanwhile, Pierre Maquet at the University of Liège in Belgium has started to use fMRI to chart the patterns of brain activity associated with different sleep states. His team is already finding that the distinctions between sleep and wakefulness look quite different when you compare patterns of activity across the whole brain, including deep structures, as opposed to using EEG, which measures activity in just the outer few millimetres of the cortex.
Lurking in the background is the hope that these approaches will shed light on the vexed issue of what sleep is for. A leading theory is that it is important for memory consolidation. Yet one of the puzzling aspects of the two Italians with status dissociatus is that, despite complete disruption of both their REM and NREM sleep, they showed no memory deficit. Does this add grist to the mill of those who believe that sleep has no other purpose than to save energy and keep us safe (New Scientist, 15 March 2008, p 30)? Or does it mean, as Mahowald believes, that the two individuals were in fact experiencing some sort of mosaic sleep? Armed with a less black-and-white definition of sleep and wakefulness, and more sensitive tools for measuring them, that question might finally be resolved.
When a crime is not a crime
In 1988, a 23-year-old Canadian man named Kenneth Parks went on trial for the murder and attempted murder of his parents-in-law. One night in the previous year, he drove 23 kilometres across town to their home and attacked them with a tyre iron and a knife. He was acquitted on all counts, on the grounds that he was sleepwalking at the time. Since then, numerous defendants have pleaded "parasomnias" - an old-fashioned term for disorders involving partial arousal during sleep that is now mostly used in law. Mark Pressman, at the Lankenau Hospital in Wynnewood, Pennsylvania, estimates that they are used as a defence in between five and 10 cases worldwide each year.
This raises tough issues about criminal responsibility and has been a source of ongoing dispute among lawyers. One of the most contentious areas concerns men charged with sexual assault who claim sleep-related sexual behaviour, or sexsomnia, as a defence (New Scientist, 28 October 2006, p 40). Sexsomnia is a variation on sleepwalking and a recognised condition. The controversy arises when these men claim - sometimes successfully - that heavy alcohol consumption provoked their sexsomnia. "My colleagues and I have been actively opposing this defence on the grounds there is no scientific evidence that severe intoxication causes sleepwalking," Pressman says. But he admits that it is a fiendishly difficult area of law, not least because no sleep study can shed light on what happened in a single individual's brain on a single night in the past.
Ground control to Major Tom
Every so often, commercial pilots and other professionals whose jobs require them to be highly vigilant have to take something called the maintenance of wakefulness test. They must sit in a comfy chair in a dimly lit room and stay awake for 40 minutes, a procedure that is repeated four times over the course of 8 hours. If they should drop off, this is detected by EEG, eye movement and jaw slackness.
However, these are fairly crude measures, and anyway people whose job is on the line tend to be highly motivated to stay awake. Sleep scientists acknowledge that a more sensitive test is required, and David Dinges at the University of Pennsylvania, Philadelphia, believes he has one.
His is a test of reaction time in which subjects have to repeatedly press a button in response to a light going on. This psychomotor vigilance test is "unbelievably sensitive" to instability in the boundaries between wakefulness and sleep, Dinges claims. A shortened version went up to the International Space Station in July, so that the astronauts could put it through its paces. If it accurately reflects their ability to do their job, it may soon be routinely applied to pilots, nuclear power station operators and others in life-or-death jobs.
Laura Spinney is a writer in Lausanne, Switzerland
