We’re not talking about Frank Lloyd Wright drafting in his sleep, but we are talking about sleep architecture. What does a typical night of sleep look like? What is a sleep cycle? How about sleep phases? We’ll try to answer these questions here.
Sleep Depth
First, let’s think about what sleep looks like over the course of the night by considering sleep depth: how easily you might be woken up (or how close are you to being awake) during a good night of sleep.
Does the graph below lead you to think: “That’s what a good night of sleep looks like”? If so…
…you’d be wrong. But you wouldn’t be alone (I’ve asked some very smart folks to draw their best guess and this is how it usually comes out). Before the advent of sleep science, this was how people thought we slept each night (with, perhaps, the exception of our astute friend, Shakespeare).
A more accurate picture of a night of sleep is this:
Notice how there’s lots of deeper seep at the beginning (that’s where you’ll find most of the aptly named “Deep sleep”), but there are also cycles of sleep. These sleep cycles occur 3 to 6 times each night and they generally last about 90-110 minutes each.
Sleep Phases
Here at Zeo, we generally talk about four sleep phases. While it’s a bit of a cheat to include Wake as sleep, it does play a role as one of the phases. Here’s a quick run through of all four:
- Wake: You start in bed awake and then fall asleep until the morning, right? Not quite. Sleep cycles get lighter several times over the course of the night, and it’s not unusual to wake up at the end of any given sleep cycle. In fact waking up a few times during the night is actually okay– it’s how the human body (and brain) was built.
- Light Sleep: Making up about 50-60% of the night, a night of sleep usually starts in Light Sleep. Light sleep can also usually be found between phases of REM and Deep Sleep. It’s uncommon to go move between REM and Deep Sleep without passing through Light Sleep.
- Deep Sleep: Making up around 10-15% of the night, Deep Sleep is the deepest sleep phase and shows up mostly in the first third of the night. When the body comes out of Deep sleep, it usually goes back to light sleep before heading into REM.
- REM Sleep: Making up about 20-25% of the night, REM Sleep usually shows up at the end of each sleep cycle. It’s more common later in the night, especially during the early morning hours. It’s also pretty common to wake up around this part of the sleep cycle.
Check out this pretty typical graph of of my own sleep. Note each of the phases and the overall time as I cycle through them:
Being a Sleep Architect
The short of it is that everyone’s sleep is different, but as we’ve seen, there are common characteristics that everyone shares. I’ll explore more of the details of sleep architecture in the future, but for now, what what does your sleep architecture generally look like? What sorts of questions does this raise for you?
Steve (ZQ: 100) is our resident Research Scientist, working to help people understand the science behind Zeo, as well as the science of sleep. He is one of the driving forces behind our Sleep Research Center.
{ 5 trackbacks }
{ 14 comments… read them below or add one }
Very interesting. I actually recently did essentially the same thing (averaged and graphed 237nights worth of 30second resolution data from the csv file downloaded from online) with the following result: Average Night.
Also, for anyone who might be interested I used a very simple python script to automate pulling the 30second resolution data from the downloadable csv file: Code.
Slightly easier and definitely quicker than copying/pasting between worksheets.
Hey Brian,
Very cool stuff. That script may come in handy around here, too.
Thanks – Steve
- Doesn’t that graph sort of remind anyone of a capacitor discharging, but where the capacitor is losing some of its total available maximum capacity over the time-span of the graph?
- It (that graph) reminds me of the question of ‘sleep pressure’. What is a scientific definition of ‘sleep pressure’? [Do any Zeo Inc.-associated Scientists know?] Maybe, is ‘sleep pressure’ related to some type of aggregate electrical ‘imbalance’, or aggregate need of nerve cells of the brain and body (body and brain) to fully ‘repolarize’?? or ‘to reachieve full (or fuller) bioelectrical balance’ [[does that - bioelectric rebalancing - involve relative concentrations between particular types of neurotransmitters? and, how would you measure the 'sleep pressure' of someone who does hard physical labor all day - would you measure how much 'muscle' cell restoration is needed - maybe that's not so 'bioelectrical', but rather is related? to the time scales of cellular 'repair'/'recovery' processes?]], or, is ‘sleep pressure’ related to something a bit different?
Is it scientifically possible to design a measure, a metric, a way of measuring ‘sleep pressure’? [Maybe before going to bed, a device could tell me 'plan on trying to get X hours of sleep this coming night, based upon your brain's-and-body's currently- measured/ estimated 'sleep pressure'?]
Also, I probably wouldn’t be satisfied with an answer to ‘what is sleep pressure’ if it didn’t explain to me, why can’t my [daily 'resting'] heart rate be its normal ‘slow’ [and healthier] resting heart rate when my body and brain have a 1 night or multi-night overall sleep deficit? It seems that its not just about the heart itself [not just about if the heart itself has had enough rest]??, but rather about some ‘background’ bioelectric state of overall body and brain bioelectric non-imbalance, which allows the heart to function more healthily.
Great comment Norman. I love the analogy of reducing sleep pressure and capacitor discharge. Never really thought of it like that before.
Sleep pressure is an effect of one of the two basic processes that regulate sleep in humans – namely Sleep Homeostasis (the other being the Circadian Rhythm). Sleep Homeostasis basically means that the longer you’re awake, the more sleepy you get. This seems to be common sense, and there are a few biological markers for increases in sleep pressure. The first is slow wave activity, which is similar to what we call Deep sleep around here. Deep sleep is dependent in large part on sleep pressure – the longer you’ve been awake, the more Deep sleep you’re gonna get (though it’s a very bad idea to stay awake for a long time just trying to get more Deep sleep). However, you have to have slept to see that marker, so what else is there? Theta activity (brain activity at around 4-7 Hz) during wake has also been proposed as a marker for sleep pressure. More theta activity has been associated with increases in sleepiness, especially caused by partial and total sleep deprivation. Some researchers have also suggested that the neurotransmitter adenosine may be a marker for sleep pressure as well, thought that’s less well established.
As for why sleep loss has an effect on heart rate? Well, sleep deprivation causes an increase in cortisol levels, which is associated with increased sympathetic activation – which can lead to increased heart rate, blood pressure, etc. Why do cortisol levels increase? I’m not so sure about that. But at a broad level, to think of it as an “imbalance” seems appropriate. That’s why it’s called “homeostasis”, and when you’re awake for long periods of time, or not getting enough sleep, you’re pushing the boundaries of an equilibrium.
I hope that helps, sorry it was long-winded, but it gets a whole lot more complicated than even that, if you’re interested to read more.
Is the fact that our deep sleep tends to be earlier in our sleep cycle why it’s recommended to start sleep before midnight? Is that actually true that we get better sleep before midnight? I thought
it was an ‘old wives’ tale but I DO feel better the few times I get to sleep by 11 or 11:30..
Zeo made it possible for me to see & determine that my well-being relies heavily on naturally acquired deep sleep (meaning sleep onset around 10pm/no meds). After only a few days my fibrofog cleared & a mild aphasia disappeared. Summer arrived & with it later to bed & later to rise. Still great sleep scores but less deep sleep. Aphasia & brain fog returned! Daytime fatigue Is also greater!
It seems to me that sleep cycles are an evolutionary result of never having too long of a period of down time. Given that we no longer need to fear being eaten by a lion during the night, I imagine that if one could engineer 90 minutes of deep sleep followed by 2 hours of REM sleep we could get a great sleep in 4 hours. Comments?
Hi, I agree .the most I had on my Zeo :
total sleep 9hrs
light sleep 1:38
deep sleep 5:30
REM 1:56
times woken 1
score that night 155
What about napping? I love my naps, and think I’ve struck a pretty good balance between 6-7 hrs/night sleep with a .5 – 1 hr nap in mid afternoon. I find them enjoyable and refreshing, and also a great tool for enhancing my creativity…
Why can’t my Zeo track my naps? It seems like the ZQ discriminates against nappers, and presumes that the only valid sleep schedule is one session per night.
Great comments all around. Nina and terri – Deep sleep does tend to occur early in the sleep episode, but doesn’t generally depend on when you go to bed. That’s according to the current understanding of sleep science… however, we have been hearing more and more about peoples’ Deep sleep being related to when they go to bed. That’s something that I will have to do more research with. Interesting topic!
Mike, that seems like a good idea, but we also have to keep in mind that Light sleep may also be very important and that sleep cycles may play a role – that consolidating sleep too much may not be advantageous. If you’re interested, here’s a good paper on the co-evolution of sleep and warm-bloodedness.
Tom, I hear your pain about Zeo’s inability to track naps properly. That’s due to a technical issue, but we are well aware of the fact that many of us would like to be able to better understand and track our naps. It’s something we talk about here at Zeo, and are working to improve. Thanks for the input!
I don’t know what to make of my sleep pattern – Zeo tells me that I fall asleep very rapidly, but then only go into REM sleep – typically much less than 5% of the time in Light or Deep sleep. This doesn’t seem to vary with what I am doing before going to sleep, etc. Is this pattern common?
I do have a sleep problem in that I fall asleep during sedentary activities, (watching TV, lectures, concerts, etc) and wonder if this might be related to my lack of light/deep sleep.
Using the Zeo, I’ve found that deep sleep is also critical for waking up and feeling refreshed as opposed to say …run over by a mack truck. But what I have found is that my deep sleep doesn’t look like other peoples graphs. It is not a solid block of time but is rather 5 minute pieces strung together across other cycles. It’s completely fractured. Here is where I lost faith in Zeo because it has no way to measure that and yet for me, it is clear a major issue.
Hi Leroy and Susan. Thanks for the contributions. Everybody’s sleep is different, to be sure, but that’s not the end of the line.
Leroy, I wonder if there might be something else going on with your readings. If the headband is moist or it’s really humid, it might be misreading a lot. I suggest you contact customer support to get more info to look into possible technical solutions.
Susan, sorry to hear about your loss of faith in Zeo. Hopefully, you can take advantage of our coaching tools and sleep information center for more information and strategies for sleep.
You guys can also check out the forum to interact with others out there who may be experiencing the same things.
Perhaps you will find interesting the following article:
Rial, R.V., et al., Evolution of wakefulness, sleep and hibernation: From reptiles to mammals. Neurosci.
Biobehav. Rev. (2010), doi:10.1016/j.neubiorev.2010.01.008
R. Rial