Lucid dreams, the kind in which dreamers become aware that they’re dreaming, often allow control of the dreams’ narratives. Unshackled from space and time that govern waking life, lucid slumberers can explore endless possibilities of the dreaming world. Lucid dreaming could also help researchers peer inside the dream state in new ways. In a study published today (February 18) in Current Biology, scientists show that lucid dreamers can process and exchange complex messages with the waking world.
In other lucid dreaming studies, sleepers have signaled lucidity with eye movements, allowing researchers to distinguish brain activity during these episodes. But to learn the content of these dreams, researchers still rely on sleepers’ recollection upon waking. “Of course, this relies on the memories of the participant, and this might be distorted,” says Kristoffer Appel, a sleep and dream researcher at Osnabrück University and the...
In a proof-of-concept study, the sleep researchers recruited volunteers who were frequent lucid dreamers or who learned lucidity-inducing techniques. While the participants snoozed, the researchers used electroencephalography (EEG) analyses of brain activity to confirm that they were asleep. In response to yes/no questions and simple math problems, six participants correctly answered a total of 29 questions with pre-arranged eye signals.
The Scientist spoke with Appel to learn more about the research on communication with lucid dreamers and the doors it could open for dream research, learning, and even entertainment.
The Scientist: What was the inspiration for doing this study?
Kristoffer Appel: First of all, it’s exciting to try something that is seemingly impossible—to communicate with someone who’s sleeping. So it’s kind of cool to try to interact with these people and see if you can actually get answers to your questions and try to see if you can find some new ways of doing dream research. So this lucid dreaming study was a step forward from what has been done previously in other lucid dreaming studies. And we also tried to continue this line of research and find a way to transfer actual messages with real content between the dream world and the waking world.
TS: Could you describe your experimental design for interacting with lucid dreamers?
KA: This study was conducted by four independent labs in four different countries, which is also kind of special. . . . So the idea was to have someone achieving lucidity and in the dream state and then interact with them either via spoken messages or via beeping tones, flashing lights, or tactile stimulation to get messages inside the dream. And also have the person answer these messages either by eye movement that you can control from within the dream or via facial muscle contractions.
TS: How did you ask questions to the dreamers?
KA: In Germany, I asked math questions via Morse code. For example, for the question, ‘three plus one,’ would be something like ‘dih dih dih diiiih diiiih’ for ‘three,’ then ‘dih diiiih diiiih dih’ for ‘plus’ then ‘dih diiiih diiiih diiiih diiiih’ for ‘one.’ These beeping tones are presented to the subject sleeping in the sleep laboratory. And they incorporate this stimulation into their dream. For example, in their dream, they are at a bus station and there’s a ticket machine and it is beeping. And they realize, ‘Okay, this is the message from the waking world and I have to understand what math problem this is.’
So they decode this in their dream and figure out, ‘It’s three plus one. Now I have to send the answer to this question.’ And they do this either by facial movements or by moving their eyes left and right. So in this case, the answer is four. So they move their eyes four times left and right, which can be recorded in the sleep laboratory. That’s how this communication in the German case worked.
Other labs, the French and American labs and also the Dutch lab, they were just using spoken language. So they asked the participant, ‘What is three plus one?’ And the participant incorporated this whole spoken text into their dream, so maybe as some sort of narrator voice like in a movie. The French team also used normal questions. For example, ‘Do you like chocolate?’
TS: What was the main finding?
KA: The main finding was that it is possible to interact with a sleeping person, to exchange messages . . . and have the answer in real time, without [the dreamer] waking up, transferred back to the sleep laboratory experimenter. And of course this has quite large implications. So this view that people are shut off from the waking world during sleep, this has to be updated. This is basically a new method for investigating and understanding dreams.
TS: What was different about this study from other research on lucid dreams?
KA: Previous lucid dreaming studies [have used] eye movements for time stamping some events in the dream. But here you have these complex questions and complex answers in which you have to maintain quite sophisticated cognitive abilities. Decoding Morse code during dreaming is not that easy. Also, just remembering the task that you are supposed to do and how to do this is much more complex than what was done before.
TS: How do you see the possibility of communicating with dreamers being used going forward?
KA: There are all sorts of applications. So in the short term, for dream research, this is an important study because we can now actually investigate the dreaming state directly and interactively. This is normal in other scientific and psychological experiments, that you can interact with your participant and ask questions and so on, but during sleep this of course is not so easy. So this is the first step in how this can be possible.
It is possible to interact with a sleeping person, to exchange messages . . . and have the answer in real time.—Kristoffer Appel, Osnabrück University and the Institute of Sleep and Dream Technologies
But also it has some real world applications maybe. For example, learning during sleep. That you can actually transfer new knowledge into sleep . . . and then still remember it after waking up. This is kind of the dream of every student—to learn the vocabulary for the Latin test during sleep. But also you could use it for psychotherapy or maybe have some new form of nightmare therapy to assist the patient with nightmares in real time and coach them somehow. Also to use it for problem solving or for creative purposes— maybe artists or writers or composers might use this to dream of works and directly transfer them to the real world without forgetting them when they wake up. And of course for entertainment. Someday we might have some sort of guided dreams where instead of reading or watching a James Bond movie, you’re actually James Bond yourself in your dream.
TS: How could interactive dreaming be used to understand dreaming?
KA: There are many different things. For example, how does dreaming work? What actually influences dreams? If there are some stimuli which get incorporated into the dream and some which don’t get incorporated into dreams. How does the narrative of a dream develop? If you can manipulate somehow the course of the dream, then you might find out more about how the brain usually does it with and without this intervention. So you could find out a lot of things about dreaming.
TS: What’s next for your research on lucid dreaming?
KA: One thing that I find very exciting is how you can actually increase message complexity. Instead of just asking yes/no questions or math questions, you might think of having answers be more complex, like reporting really what you’re seeing in your dream. You might be able to report things like, ‘I see a house,’ ‘I see faces,’ ‘I see the beach,’ ‘I’m terrified,’ ‘I’m happy.’ There are many things that you might want to communicate which go behind beyond yes/no. So I’m working on a method for that, which uses more sophisticated eye movements for transferring words and sentences to the waking world.
K.R. Konkoly et al., “Real-time dialog between experimenters and dreamers during REM sleep,” Curr Biol, doi:10.1016/j.cub.2021.01.026, 2021.
Editor’s note: The interview was edited for brevity.