Monday 17 June 2019

Sleep talk: what does it mean?

While being a harmless disorder, it might reveal our hidden depths, writes Shane Cochrane

'Sleep talking is a sleep disorder'
'Sleep talking is a sleep disorder'

Back in 2015, an American lady began posting on Twitter the things her boyfriend said in his sleep. He said a lot of strange things.

"I don't care what you say, I'm getting tigers."

"Did you cut off the leg?" "Your eyeballs, they smell like eyeballs."

"I'm not braking for ducks."

"Why is it always clowns?"

By 2016, he'd become an online celebrity - albeit an unwitting one. The really strange thing is - he's far from unique. There's a good chance that many of us will talk in our sleep like this at some point in our lives.

But is it just a strange and entertaining quirk, or is it possibly a window into the workings of our minds?


Sleep talking is a sleep disorder, like sleepwalking or night terrors, which researchers at the Sapienza University of Rome have described as "the utterance of speech or other psychologically meaningful sounds during sleep, without simultaneous and subjective awareness of the event." And while sleep talking is just one of the family of "sleep utterances" - such as mumbling, groaning, whistling and laughing - they say that "the psychologically meaningful quality of the event is the discriminating feature."

Most utterances are made up of a few words and last only one or two seconds. They can be spoken, whispered, or shouted, and range from "meaningless sequences of words" to "long and articulated phrases".

Sleep talking episodes can happen at any time of the night, but are more likely to occur during the first three hours of sleep. They can take place during both the REM (rapid eye movement) and non-REM stages of sleep, but episodes during REM sleep tend to be more emotional than those in non-REM sleep.

As most people are unaware of their night-time ramblings, researchers have found it difficult to estimate the prevalence of the phenomenon. However, a recent French study suggests that as many as 67pc of us will talk in our sleep at least once in our lives, making sleep talking the most common sleep disorder.

Both men and women can experience episodes of sleep talking, but it's most common in children, adolescents and young adults.

The disorder is predominantly genetic, but it can also be triggered by environmental factors. "Sleep talking shares this aspect with most parasomnias," says Professor Luigi De Gennaro of Sapienza University. "Environmental factors trigger episodes of sleep talking or increase their frequency. The most common factors are emotional stress, febrile illness, alcohol, certain medications or sleep deprivation itself. These factors also act as a trigger for other parasomnias, like sleepwalking."

Sleep talking can occur on its own or alongside other sleep disorders. It can also occur alongside other medical conditions, especially psychiatric conditions, such as post-traumatic stress disorder.

While Professor De Gennaro warns that the sudden appearance of this sleep disorder "can be considered as a potential sign that the individual has been exposed to some kind of physical or psychological stress", sleep talking is generally harmless and may disappear spontaneously.


Research into the disorder has been sporadic, and most studies were done quite some years ago. The condition's reputation for having few health consequences is a big reason why it has received little scientific or medical attention. But it's also been neglected because it's proven to be a very difficult phenomenon to study.

Professor De Gennaro says: "After the first studies in the 1960s, describing sleep talking from a qualitative and phenomenological perspective, further knowledge was precluded by the lack of available measures of the associated brain activity. As an example, neuroimaging techniques like fMRI are not viable. If you try to use them, sleep talking stops or the subjects are awakened."

And as sleep-talking episodes are quite rare, a certain amount of luck is also needed in recording them. For example, in one study, a relatively prolific sleep talker had to be monitored for four days before the researchers were able to capture anything.

And in Professor De Gennaro's current project, his pool of 1,309 sleep-talking candidates quickly became 50. And by the time the study began, only 13 were found to be viable. And out of that group, only a few had sleep-talking episodes during the study. "Sleep talking is an erratic phenomenon," he says.


Despite these difficulties, Professor De Gennaro believes that by studying sleep-talking we can learn a lot about how our brains work.

One area where he believes sleep talking can help is dream research. Dreams are a difficult thing to study: they're not directly observable, and so researchers are reliant on the dreamer being able to accurately describe their dreams.

And while these "dream reports" can be quite reliable, it's known that certain elements of a dream are easier to recall, while others can be omitted or even distorted by the dreamer.

But given that a number of studies have found that there can be a high degree of concordance between recalled dreams and "speech episodes" recorded during those dreams, Professor De Gennaro believes that sleep talking may provide a way of capturing dreams without distorting them.

"Although empirical evidence is scarce, sleep talking tends to be consistent with the content of dreams," he explains. "The existence of this phenomenon could represent a potential way to directly access dreams."

Access to our dreams could be an important step in better understanding how we learn, or as Professor De Gennaro's team explain it in their research paper: "The role of dreaming in learning processes provides evidence that dreams could be possibly considered as a sort of 'cognitive' replay occurring during sleep.

"Given this assumption, it appears clear that sleep talking episodes, mirroring dream content, could represent a window on the cognitive processes related to the incorporation and consolidation of memories during sleep."

Professor De Gennaro also believes that studying sleep talking can potentially help us better understand cognitive functions, such as sleep and speech production.

For example, Dr Anastasia Mangiaruga, a PhD student working in Professor De Gennaro's sleep lab, is currently carrying out a study that involves monitoring electrical activity in the brain during sleep-talking episodes. Her early findings suggest that the parts of the brain that are active during sleep talking are the same parts of the brain that are involved in speech production when we're awake.

This kind of insight may prove useful in the study and treatment of speech disorders, such as the aphasia experienced by stroke victims.


So, what do we actually say when we talk in our sleep?

To find out, researchers, led by Dr Isabelle Arnulf of the Pitié Salpêtrière Hospital of the Sorbonne University, Paris, put 232 adult sleep talkers in a laboratory for two consecutive nights. In addition to monitoring their brain waves, oxygen levels, breathing, heart rates, and eye movements, like in any other sleep disorder study, the researchers made video and audio recordings of the sleepers.

They then noted every verbal utterance - including talking, crying, shouting, whistling, mumbling and whispering, as well as any silent lip movements - made on the recordings.

The team recorded 883 utterances, of which only 361 could be described as "speech episodes," meaning clearly discernible individual words and sentences. The rest were non-verbal utterances, such as mumbles, groans and cries. Of the "speech episodes," 22pc contained profanities, with the 'F' word featuring quite prominently. In fact, only 12 of the 361 "speech episodes" contained anything close to polite language.

But regardless of whether or not they were swearing, all of the sleepers' sentences were grammatically correct. The sleep talkers said "no" a lot. It was by far the most frequently spoken word. A couple of them spoke in foreign languages. One sang a song. And a man who stammered while awake also stammered in his sleep - which was very unexpected.

But what was really curious was that 90pc of what was said was directed at "an invisible third person." "Interestingly, when the sleepers spoke with one or several persons, they left an appropriate silence, as if listening to a response from their fictive talker," the researchers noted. "The gaps between turn-taking were respected after questions and orders, as if speaking awake. This means that the dreamers spoke only those words that, in the dream, they experienced as their own." Just as interesting was the researchers' observation that only 10pc of what was said by the sleepers could be construed as "inner speech," leading them to conclude that "sleep talking does not unmask our inner silent verbal thinking."

And if you're curious about your own night-time utterances, there are consumer apps available, like Sleep Talk Recorder and Dream Talk Recorder.

Health & Living

Editors Choice

Also in Life