Sleep: What Actually Happens When You Get Too Little — or Too Much

We spend roughly a third of our lives asleep. For most of human history, that fact was treated as a slightly embarrassing inefficiency — a biological tax on the productive hours. Thomas Edison famously called sleep “a criminal waste of time” and claimed to get by on four hours a night.

The last few decades of sleep research have quietly demolished that view. Sleep turns out not to be the absence of activity but a different kind of activity: a scheduled, structured, metabolically expensive process the brain and body cannot skip without paying for it somewhere else.

Here’s what the evidence actually shows — including the parts that are less tidy than the headlines suggest.

What Sleep Is Actually Doing

A night of sleep is not a flat line. It cycles roughly every 90 minutes through distinct stages, each with its own job.

Deep sleep (slow-wave sleep) dominates the first half of the night. This is when growth hormone is released in its largest pulse, when tissue repair happens, and when the brain appears to do its housekeeping. Research published in Science in 2013 described a mechanism — the glymphatic system — in which the spaces between brain cells expand during sleep, allowing cerebrospinal fluid to flush through and clear metabolic waste products that accumulate during waking hours. Among the substances cleared is beta-amyloid, the protein that aggregates in Alzheimer’s disease. This is still an active research area and the human evidence is less complete than the mouse evidence, but the basic finding — that the sleeping brain cleans itself in a way the waking brain does not — has held up.

REM sleep clusters in the second half of the night and is when most vivid dreaming occurs. It appears to be central to emotional processing and to certain kinds of memory consolidation, particularly the integration of new information into existing knowledge. There’s a reason “sleep on it” is folk wisdom in almost every language.

Both matter. And crucially, they aren’t distributed evenly — which is why cutting your night short by two hours doesn’t remove a uniform 25% of each. It disproportionately amputates REM sleep, because REM is loaded toward the morning.

What Happens With Too Little

The immediate effects

The most reliably demonstrated consequence of sleep restriction is cognitive. In a well-known 2003 study, researchers restricted participants to 4, 6, or 8 hours in bed for two weeks and tested them daily. The 6-hour group — a schedule millions of people consider normal and sustainable — showed cognitive deficits after two weeks that were comparable to going 24 hours without sleep at all.

The unsettling part wasn’t the deficit. It was that participants in the restricted groups reported feeling only slightly sleepy. Their subjective sense of impairment plateaued while their actual performance kept declining. People are poor judges of their own sleep-deprived state, which is precisely what makes chronic short sleep so easy to sustain.

Reaction time, attention, and error rates degrade first. Judgment about risk appears to degrade alongside them — a combination that explains why drowsy driving is treated by transport safety bodies as a hazard on par with drink-driving.

The metabolic effects

Short sleep shifts appetite hormones. Ghrelin (which signals hunger) rises; leptin (which signals fullness) falls. Insulin sensitivity drops measurably even after a few nights of restriction in otherwise healthy young adults. Large observational studies consistently link habitual short sleep to higher rates of obesity and type 2 diabetes.

The caveat matters: these are largely correlational. Poor sleep and metabolic dysfunction reinforce one another, and untangling which came first is genuinely difficult. But the short-term experimental work — where you take healthy people and restrict their sleep — does show the hormonal shifts happening in real time. The mechanism is plausible, not just the correlation.

Immune function

One of the more elegant experiments in this area: researchers exposed participants to a rhinovirus (the common cold) via nasal drops after tracking their sleep for a week. Those sleeping less than six hours a night were markedly more likely to actually develop a cold than those sleeping seven or more — despite everyone receiving the same viral dose. Similar work has shown reduced antibody response to vaccination in sleep-deprived individuals.

Cardiovascular risk

Chronic short sleep is associated with elevated blood pressure, increased inflammatory markers, and higher rates of cardiovascular events in large cohort studies. There’s even a quasi-natural experiment: several analyses of the spring transition to daylight saving time — where an entire population loses one hour on a single night — have reported a small but detectable uptick in heart attacks in the days that follow. The effect size is debated and some replications are weaker than others, but it’s a striking illustration of how a population-wide sleep perturbation can show up in hospital data.

What Happens With Too Much

Here is where the picture gets genuinely more interesting — and where most popular coverage goes wrong.

Large epidemiological studies consistently find a U-shaped curve. Mortality and disease risk are elevated at the short end (under six hours) — and also at the long end (typically over nine hours). Long sleepers show associations with cardiovascular disease, depression, cognitive decline, and all-cause mortality that are, in some datasets, as strong as those for short sleepers.

This gets reported as “too much sleep is bad for you.” That interpretation is probably wrong, or at least badly oversimplified.

The reverse causation problem

The far more likely explanation for most of the long-sleep association is that long sleep is a symptom, not a cause. Consider what makes someone sleep ten hours a night:

• Undiagnosed sleep apnoea — where sleep is fragmented and non-restorative, so more time in bed is needed to get less actual rest
• Depression, which frequently presents with hypersomnia
• Chronic inflammatory conditions
• Undiagnosed illness, including some cancers, in the years before diagnosis
• Certain medications

In other words, a great deal of long sleep is the body’s response to something being wrong. Studies that carefully control for underlying health conditions tend to see the long-sleep risk shrink substantially — though notably, it doesn’t always vanish entirely, which leaves the door open for a genuine effect that nobody has fully explained yet.

What long sleep does reliably do

Two things are on firmer ground.

First, sleep inertia. Waking from deep sleep — which is more likely if you’ve slept an unusually long time and your final cycle is out of its normal position — produces that heavy, foggy, disoriented feeling that can persist for 30 minutes or longer. This is why a very long lie-in can leave you feeling worse than a normal night. It’s real, but it’s a transient state, not a health consequence.

Second, circadian disruption. Sleeping until noon on Saturday and Sunday after waking at 6am all week creates what researchers call “social jetlag” — a mismatch between your body clock and your schedule equivalent to flying across several time zones every weekend. This is associated with metabolic and mood consequences, and it’s a mechanism where the causation is far more plausible in the direction people assume.

Which suggests something useful: irregularity may matter as much as duration. Recent work using wearable data has found that sleep regularity — going to bed and waking at consistent times — predicts mortality risk at least as well as total sleep duration, and in some analyses better. That’s a relatively new finding and worth watching, but it points at something intuitive: the body runs on a clock, and the clock likes to be told the same thing every day.

So How Much Is Right?

The consensus recommendation from bodies like the American Academy of Sleep Medicine and the NHS lands at seven to nine hours for most adults. Older adults often need slightly less, teenagers considerably more (and are biologically inclined toward a later chronotype, which is why early school start times are such a poorly designed fight against adolescent physiology).

But the population average conceals real individual variation. A small number of people carry rare genetic variants — in genes including DEC2 and ADRB1 — that allow them to function normally on six hours or fewer with no apparent cost. Estimates suggest this is well under 1% of the population. Which means that if you believe you’re a natural short sleeper, the statistics say you’re almost certainly wrong. Most self-identified short sleepers are simply chronically sleep-deprived people who have normalised the feeling.

The more useful test isn’t the number on the clock. It’s whether you wake without an alarm feeling reasonably restored, and whether you can get through mid-afternoon without a crash. If you sleep dramatically longer on weekends than weekdays, that gap is your sleep debt, and it’s telling you something.

The Honest Summary

Sleep deprivation is one of the better-established risk factors in modern health science — the experimental evidence is strong, the mechanisms are increasingly well understood, and the effects show up quickly.

Oversleeping is a murkier story, and the murkiness is worth respecting rather than papering over. The correlations are real, but the causal arrow probably points backwards for most of them: long sleep is more often a signal that something needs investigating than a problem in itself. If you’ve suddenly started needing ten hours a night when eight used to do, that’s a reason to see a doctor — not a reason to set an alarm.

The one thing both ends of the curve have in common is that the body appears to want consistency. It runs on a clock. The evidence increasingly suggests the clock cares less about being generous with it than about being predictable.

---

This article is for general information and is not medical advice. Persistent sleep problems — including chronic difficulty falling asleep, unrefreshing sleep, loud snoring, or a sudden change in how much sleep you need — are worth discussing with a GP or healthcare professional, as they can indicate treatable underlying conditions such as sleep apnoea.