A new explanation for why we have nightmares contradicts the conventional wisdom about sleep

Sweet dreams.
Sweet dreams.
Image: AP Photo/Jacquelyn Martin
We may earn a commission from links on this page.

The predominant thinking has been that most of us get too little sleep, and that we should all be worried about the loss of productivity, dips in mood, and buildup of waste proteins in our brains that are a result of staying up too late.

But there may be a frightening side effect of getting too much sleep. In a survey of 846 adults in the UK, researchers from the University of Oxford found that those reporting more than nine hours of sleep a night were 40% more likely to have nightmares than those who either slept for shorter amounts of time.

Stephanie Rek, a neuropsychologist and lead author of the study, and her team passively recruited participants through advertisements on social media and contacting individuals listed on existing databases of people who said they’d be interested in sleep studies. Willing adults filled out a survey online reporting their usual amounts of sleep, how often they drink, and how often they worry about things—especially before bed. They also reported how many nightmares they had over a period of the last two weeks.

Of the 400 or so participants who said they had nightmares, those who reported worrying the most or getting more than nine hours of sleep were more likely to have nightmares, according to New Scientist. The paper suggests this is probably because the longer you sleep, the more opportunity there is for you to have bad dreams. Nightmares, like any dream, occur during our deepest sleep during the REM phase, which is one of four parts of a sleep cycle we go through during the night. People who sleep more are going to have more REM sleep.

Less surprisingly, the study also found that those who reported worrying about the future or failure prior to falling asleep also were more likely to have nightmares than those who reported being generally less anxious.

Like any survey-based study, this comes with caveats: For one thing, researchers couldn’t control who responded; they ended up with a sample size that was overwhelmingly white and female with an average age of 44—not exactly representative of the entire population. And any survey that relies on self-reported data that could be limited by participants’ faulty memories. Finally, surveys only present a limited swatch of data and can’t prove causality between sleep duration, fretting about the future, and nightmares.

The good news is, within this survey at least, drinking wasn’t significantly tied to bad dreams. You may as well have a nightcap to that.