Home » Understanding Sleep » Basics of Sleep » Hypnograms – Graphs Representing Your Sleep

Hypnograms – Graphs Representing Your Sleep


Sleepline.com has spent thousands of hours researching and comparing products relating to sleep. We are serious about recommending the correct products for your individual needs. We take in more data points than just our own and encourage our users feedback. At times we are compensated for the links you click at no cost to you. Learn more on our disclosure page here.

Best way to chart your own sleep

Apple Watch Series 6 – When Apple dropped the Series 6 version of their popular smartwatch, they mentioned that sleep tracking would be a major feature. They weren’t kidding. Apple Watch includes a robust suite of sleep-enhancing features including goal-setting, sleep stage tracking, a wind down function (to help with your pre-bedtime routine), and a soothing alarm to wake you.

Buy: Apple Watch Series 6, $384.99+ from Amazon

A hypnogram is simply a graph representation of one’s sleep cycle, as well as of nightly awakenings. It is useful when a rough image is needed so that researchers can easily visualize what someone’s sleep looks like and compare it to that of other people.

Hypnograms can be used in sleep research as well as in sleep medicine, for patients who may have a sleep disorder. Hypnograms can show all sleep stages with waking periods, and we can draw some conclusions about sleep quality, as well as monitor sleep onset and total sleep time.

What is a hypnogram and how does it work? Explained

A hypnogram is one of the ways in which sleep is recorded in the study of sleep (polysomnography). It’s used both, for research and for diagnostics in medicine. When you look at a hypnogram, you’ll see a not too detailed sleep presentation.

Figure 1. A typical hypnogram.

The reason behind it is that a hypnogram records activity every 30 seconds. Therefore, we see the macrostructure of our sleep, whereas the microstructure can be seen on other types of recordings, for example, on EEG (electroencephalogram), which clearly shows even the quickest brain waves.

In Figure 1, we can see the sleep stages during eight hours of sleep (the horizontal axis shows time and vertical sleep stages). Non-REM 1 is the lightest stage of sleep, Non-REM 2 is a deeper stage of light sleep, while, as the value goes down in the graph, it represents deeper sleep stages.

What used to be called Non-REM 3 and Non-REM 4 is today combined in Non-REM 3 (deep sleep, or slow-wave sleep).

Even though REM (rapid eye movement) usually comes after sleep stage 2, it is placed higher on the graph because the brain activity is similar to that of wakefulness. Also, many people experience really short awakenings from REM sleep – some scientists even theorize that the purpose of REM sleep is to prepare the body for waking up.

How is a hypnogram obtained?

A hypnogram is usually made at the same time as other recordings, like an electroencephalogram (EEGs – they are recordings of brain-wave activity), electrooculography (EOGs – they show eye-muscle movement) and electromyography (EMGs – they show the activity of skeletal muscles). All of the three are electrical signals obtained by electrodes placed on the skin.  

While one or more of these recordings are being made, a computer also takes information every 30 seconds to make a hypnogram.

What do hypnograms look like? Normal vs disrupted sleep

In Figure 1, we can see a series of normal sleep cycles. In the first half of the night, there is plenty of deep sleep, whereas, in the second half, there’s more REM sleep and significantly less deep sleep. There are only a couple of awakenings, and they occur after REM sleep. Light sleep precedes and follows both deep and REM sleep. This is what the typical sleep of a relatively young, healthy adult looks like.

Frequent awakenings and OSA

In Figure 2, the first hypnogram was obtained from a person who reported no frequent wakings, whereas the second one was obtained from a person who complained about sleep fragmentation. There isn’t much difference in the amount of REM sleep each of them received, but there is a big difference in how much deep, restorative sleep each of them had.

Deep sleep is ‘sensitive’ and any disorder which causes frequent awakenings also ruins the healthy sleep architecture, allowing more time for light sleep and not enough for deep sleep.

Patients with obstructive sleep apnea (OSA) rarely get a good night’s sleep because their narrow airways easily get obstructed at night, completely stopping their breathing. As they wake up to change the position, their sleep cycle ‘resets’ – getting them away from the deep sleep. One of its dangers is that OSA patients are rarely aware of these short arousals, thinking they sleep well, when in fact, all they are getting is non-restorative sleep with all its consequences.

Figure 2. A normal hypnogram and a hypnogram of a person who had frequent awakenings. Image source: Berry, R. R, Fundamentals of Sleep Medicine, 2012.


Narcolepsy is a sleep disorder associated with REM sleep, excessive daytime sleepiness, and involuntary sleep episodes. One of the characteristics of narcolepsy is that the person might fall directly into REM sleep, or even begin dreaming before sleep onset.

Figure 3 demonstrates how much REM a normal person gets between midnight and 8 am, versus REM episodes of a narcoleptic (notice that they occur at random times of the day).

Narcolepsy is usually a consequence of extreme sleep deprivation, alcoholism, or other unhealthy behaviors.

Figure 3. Normal and narcoleptic hypnogram. Image source: Berry, R. R, Fundamentals of Sleep Medicine, 2012.

Hypnograms and medications

Hypnograms can show how certain medication affects sleep architecture. For example, a pilot study from 2003 used hypnograms to test the effect of antiepileptic medications on sleep. They found that carbamazepine didn’t make any change, whereas phenytoin increased the lightest sleep stage (stage 1), while deep and REM sleep were decreased. Slow-wave (deep) sleep has shown an increase when gabapentin was administered. Therefore, some medications can be good, some detrimental, and some neutral on sleep structure of humans.

Good sides and drawbacks

Hypnograms are very useful if we want to track someone’s sleep cycles, and are not interested in the minute details. We can easily see and show others whether a person has an overall healthy sleep architecture or not. Even eager amateurs can use them and learn about sleep architecture. Many studies rely on hypnograms to track sleep stages or awakenings. However, as it captures information every 30 seconds, a hypnogram is not a very precise recording.

Using EEG allows us to see the brainwaves as they occur. For example, a hypnogram can’t show the bursts of brainwaves (sleep spindles) and individual waves called K-complexes. Sleep spindles are associated with light and deep sleep as well as memory consolidation, but they also serve in keeping the brain asleep, ‘distorting’ the outside stimuli, which usually prompt a K-complex type of the waves. Sleep spindles can show desirable brain activity, our chronotype, as well as insomnia and other problems.

If one wants to learn about brain activity in such depth, they will not use hypnograms.

Additional resources

  1. Berry, R. R, Chapter 6 – Sleep Architecture Parameters, Normal Sleep, and Sleep Loss. Fundamentals of Sleep Medicine. Pages 79-90. 2012 https://www.sciencedirect.com/science/article/pii/B9781437703269000063 Accessed March 17, 2019.
  2. Berry, R. R, Chapter 24 – Hypersomnias of Central Origin. Fundamentals of Sleep Medicine. Pages 451-479. 2012 https://www.sciencedirect.com/science/article/pii/B9781437703269000245 Accessed March 17, 2019.
  3. Swihart B. J, Caffo B, Bandeen-Roche K and Punjabi N. M. Characterizing Sleep Structure Using the Hypnogram. Journal of Clinical Sleep Medicine. August 2008. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2542492/ Accessed March 17, 2019.
  4. Swihart B. J, Punjabi N. M, and Crainiceanu C. M. Modeling sleep fragmentation in sleep hypnograms: An instance of fast, scalable discrete-state, discrete-time analyses. Computational statistics & data analysis. September 2015. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4865264/ Accessed March 17, 2019.