We spend almost 1/3 of our lives asleep. So sleep must be doing something critical, right? But what would happen if you didn’t sleep at all? The prevailing idea is that you would die – but there’s really no hard data to support this, because it’s practically impossible to deprive someone of all sleep. The official record for human sleep deprivation is 11+ days, by Randy Gardner, who was being overseen by a medical team that included sleep researcher William Dement. But even in that case, it’s likely that Randy was getting “microsleeps,” brief periods of sleep leaching into normal wakefulness. Randy survived his self-inflicted ordeal, and was even able to give a cogent press conference at its termination. And when he finally went to sleep, Randy certainly took advantage of his time in bed, but never came close to recouping the approximately 80+ hours of sleep that he had lost. So does all this mean that sleep isn’t really important for survival?
Counterarguments come from studies by Allen Rechtschaffen and others who performed long-term sleep deprivation studies in rats, and found that rats died after a few weeks of persistent deprivation. The animals developed a syndrome involving skin lesions and thermoregulation issues, but the cause of death was still difficult to discern. And this study, like many others that have utilized sleep deprivation to uncover the functions of sleep, was confounded by the many side effects of sleep deprivation – stress, additional movement, altered food consumption, and in this case, repeated dunking in water. So was it lack of sleep that killed the rats, or was it actually a sleep deprivation-associated epiphenomenon?
With this backdrop, a recent study from the lab of Giorgio Gilestro attempted to address whether sleep truly has a vital function, in this case using the fruit fly Drosophila melanogaster. They developed a technical tool that allowed the researchers to video-monitor individual flies with very high resolution in real time, and then spin each individual fly’s tube when they were inactive for a set amount of time. The typical metric for movement in flies is for flies to break an infrared beam in the center of a 65 mm tube. And sleep is defined as 5 minutes or more without a single beam-cross. In the new study, the high-resolution video showed that flies, especially females, actually engaged in in-place movements that the authors termed micromovements during periods that normally would have been categorized as sleep. By their new measurements, while there was a large range in total sleep times, some female flies barely slept at all (as little as 4 minutes a day, in one individual). And yet those flies seemed otherwise healthy. Secondly, the authors performed their high-tech sleep-deprivation experiment for practically the entire lifespan of the flies, using a 20-second immobility trigger to initiate spinning of a fly’s tube. This meant practically that a fly could rest undisturbed for 20 seconds, but then would be shaken, presumably to the point of awakening. While this approach was quite effective at keeping flies from having periods of complete immobility (not even micromovements) lasting longer than 5 minutes, the typical threshold for sleep that was mentioned above, it did not result in immediate death of the flies. Female flies did die about 10% sooner (37.5 days vs. 41 days) when sleep deprived than when left to their own sleep schedules, but males’ lifespans were not significantly affected.
From these findings, the authors provocatively concluded that most sleep is not critical for survival, at least in flies. However, other explanations should be considered as well. For example, the authors state that the newly quantified micromovements included grooming, egg-laying, and feeding, all of which should clearly disqualify a categorization of sleep (unless flies sleep-groom). But would a random leg-twitch also disqualify sleep? And what if a fly is completely immobile for 4 minutes and 59 seconds, then takes one sip of food, and then is immobile for another 4 minutes and 59 seconds? Is it accurate to say that that the fly is not sleeping at all during that 10-minute period? The point is that, while the new fine-scale resolution of analysis definitely represents an improved level of analysis, it may also be leading to an underestimation of true sleep time. So those female flies who the researchers claim are only sleeping a few minutes a day might be getting substantially more shut-eye than it seems. And in the sleep deprivation studies, it’s clear that shaking a fly potentially every 20 seconds will disrupt its normal sleep patterns. But evidence from a variety of species shows that as sleep drive builds up, animals find a way to get sleep, one way or another. Mice being sleep-deprived in a columnar drum with a rotating central wall get bumped by the wall, race ahead to the wall ahead of them, curl up to grab a few seconds of rest before the wall comes around to bump them again. People who have built up sleep debt can fall asleep at the drop of a hat in practically any situation, no matter how inopportune – sitting in the first row in class, reading a bedtime book to their child, or, tragically, even while driving 70mph on the highway. So it’s likely that fruit flies who are chronically being shaken every 20 seconds are spending a large portion of the intervening 20 seconds sleeping, despite the repeated disruptions. So maybe it’s not so surprising that flies tolerated this sleep deprivation paradigm fairly well. It’s also worth pointing out that an individual fly in a tube with a plug of food in one end has literally no competition. Sleep might not affect the ability of a fly to wander a few body lengths and slurp down a meal, but it might be critical for more complex survival behaviors needed in the wild – to name a few: avoiding predators, finding distant food sources, locating and successfully wooing satisfactory mates, and choosing optimal sites for egg-laying. Thus, a small effect on lifespan in the laboratory might be magnified into obliteration under the additional pressures of the real world.
Despite these critiques, the recent study does highlight the fact that we are still without definitive proof that sleep is vitally important, and raises the issue that with improved technology for fly monitoring, Drosophila researchers may need to revisit the definitions of sleep that we have been working with for close to 20 years. Should 5 minutes of inactivity still be the benchmark? Should a single micromovement disqualify sleep? Or do our sleep-categorizing algorithms have to grow more complex to accommodate the growing complexity of our data acquisition? The researchers’ work also points out some very interesting and as yet unexplained differences between male and female flies. Males sleep much more than females, whether or not micromovements are taken into account, and yet they appear more resistant to long-term sleep disruption? Is this because males’ greater sleep drive allows them to catnap during their 20-second rest times, whereas females can’t do so to the same degree? In naturally sleeping flies, sleep time (taking micromovements into account) was positively related to lifespan in male flies, but much less so in female flies. This again points at larger questions about how and why sleep differs, both in quality and in function, between male and female animals.
Like a restless fly, a sleep researcher’s work is never done…