How your smartphone can actually improve your sleep

Smartphones are often responsible for keeping us up late at night, checking emails or social media, but they can also be used to help us get a good night’s sleep.

There are plenty of smartphone apps on the market designed to monitor your sleep patterns. The apps record when you go to sleep and wake, and some also use your phone’s microphone and accelerometer to record snoring, talking in your sleep and other noises as well as tossing and turning.

Now computer scientists and clinical psychologists at Brown University have gone one step further with a groundbreaking new app dubbed SleepCoacher. It uses sleep analytics to generate personalised recommendations informed by scientific literature and then guides you through a self-experimentation framework to help you find the recommendations that work best for you.

11 bad habits that are ruining your sleep

“We wanted to help people go beyond just how many hours sleep they had. The app collects data about you and after a few days suggests how to improve sleep,” says lead researcher Nediyana Daskalova a doctoral student in computer science at Brown University.

Daskalova presented a paper on Sleepcoacher at the User Interface Software and Technology Symposium in Tokyo at the end of last year. She shared the results of two small pilot studies on SleepCoacher users, which found that 80 per cent of those who followed the apps recommendations at least 60 per cent of the time reported an improvement in their sleep.

How sugar affects sleep duration

The pilot studies made used of a slightly modified version of a commercially available app called Sleep as Android.

“Sleep as Android is one of the most popular apps out there with more than 10 million downloads and it has a lot of useful features. The developers were very helpful to us. What SleepCoacher does which they don’t do is find causal, correlational links and give recommendations,” says Daskalova.

The app was modified, so that in addition to monitoring sleep patterns, it also allowed users to rate how refreshed they felt in the morning, as well as noting other factors that may affect sleep. For example whether they had showered before bed, if they drank coffee or alcohol some hours before, whether they were working or stressed, had meditated or were sleeping with a dog on their bed.

“The whole idea is that you can enter whatever you think is relevant so that you are in effect doing an experiment on yourself. If you don’t get a correlation then it’s not relevant whether it’s food, exercise or something else,” says Daskalova.

Sleep-deprived and internet-mad, Hongkongers place last in healthy living survey of Asia

Using that data, the SleepCoacher algorithm is able to determine what factors, either detected by the app or reported by the user, were correlated with three key outcomes: how long it took people to fall asleep, how many times they woke during the night and how fresh they reported feeling in the morning.

If a strong correlation is detected, the algorithm generates a recommendation based on a collection of 117 recommendation templates developed by a group of clinical psychologists and psychiatrists at Brown University’s Alpert Medical School.

The participants were then sent a text recommendation such as: “On average, you go to bed around 11:52pm. We’ve noticed that when you sleep around 11pm, you are on average more refreshed. For the next six days, try going to bed around 11pm to feel more refreshed.”

SleepCoacher then guides the user through a series of mini-experiments, instructing them to follow the recommendation for several nights and then to ignore it. By measuring the difference in outcome when the recommendation is followed and when it’s not, the app determines whether the recommendation is useful for the user.

“The self-experiments help us build smarter recommendations that, over time, learn what might be actually important for individual users,” says Daskalova.

The pilot participants were all college students and Daskalova says she was surprised not only by how little sleep many of them had, but by the variation among them.

“Some people who went to bed late reported feeling tired the next day, while others reported feeling refreshed. Ambient noise at night was a problem for some people and not for others. There’s a lot of variation which is what makes this such an interesting problem to solve,” she says.

Daskalova and her team developed SleepCoacher under the direction of Jeff Huang, an assistant professor of computer science and at Brown University and leader of Brown’s Human-Computer Interaction Group.

“Our work is the first of its kind to guide people to figure out whether the data is casual, instead of just correlation,” says Huang.

“That’s particularly exciting for me. We have an approach that could work in the long term to continuously improve sleep over months, or even years,” he says.

SleepCoacher is expected to be available for free download soon – you can sign up to be notified when it’s available at http://sleep.cs.brown.edu. When you download the app you will have the chance to opt into a voluntary sleep study.

“We will encrypt the data so that your information will be private. And as an added incentive for those taking part in the study, we will give them access to the data when it comes available,” says Daskalova.

She hopes that many of the people who download the app will opt into the study, giving the researchers plenty of data to work with. The study based around SleepCoacher findings will run for 10 years, allowing users to optimise their sleep behaviour over a long series of trials and analyses.

Sleep is worth taking seriously: if you get a good night’s rest you will not only feel great the next day, your mind will be sharp and your reactions swift. A bad night’s sleep and will leave you groggy and muddle-headed the next day and long-term insomnia can lead to serious health conditions such as obesity, high blood pressure, heart disease, diabetes and depression. There are dozens of sleep apps on the market – many free and others for a small fee – designed to help you relax and develop healthier sleep habits. Here are three of the most popular ones.

Sleep

This was nominated as one of the 10 best insomnia apps by online health site Healthline. It uses a selection of relaxing music to help you fall asleep and wake in the morning. You can mix and match the app’s sounds with music from your own collection.

Available for: iOS

Sleepbot

This free sleep app allows users to set alarms and create auto-settings to get the most out of your sleep. It tracks sleep cycles and records sound levels and produces detailed tables of your sleep history –  sleep time, wake time, hours of sleep and sleep debt or hours lost. You can view and share your sleep stats from your phone.

Available for: iOS, Android

Sleep as Android

This app uses your phone’s accelerometer to work out when you are asleep or awake by tracking your muscle movements. It has a smart alarm clock with sleep cycle tracking that wakes you gently at the optimal moment for a pleasant start to the day.

The unlocked version is available for a one-off payment of US$7.

Available for: Android