Scientists building technology to record dreams
Posted On May 12, 2019
the 1991 apocalyptic sci-fi film Until the End of the World, director Wim Wenders creates a vision of the future that, in a few ways, manages to mirror aspects of life in 2019. In the film, a virtual reality (VR) headset-like device is developed that gives blind people a form of sight. But as the film goes on, the device is transformed into a tool that records dreams, and a darker side of humanity comes to light: People become addicted to their subconscious storylines, spending all their time sleeping and waking only to watch their dreams.
The device looks nearly identical to an Oculus headset, but there are more similarities between our current reality and Wenders’ 30-year-old fantasy than that alone. Today, a team of researchers from universities across the country is working on a project dedicated to figuring out how to recreate our nighttime lives on the screen. And like Until the End of the World, it’s unclear what this technology might be used for.
The as-of-yet-unnamed project is the brainchild of independent dream researcher Daniel Oldis and represents a continuation of a question he and others in his field began exploring decades ago: Are dreams learning experiences, and can behaviors in waking life be influenced by what people dream about? Oldis and others concluded that the answer must be yes based on one key point: When people dream, their bodies move and react.
“Your eyes, the muscles in your body, it’s just like you’re awake. You’re responding to the dream story,” Oldis says. “Like in waking life, the body is learning from different experiences. We learn from dreams, and it affects our personality.”
The debate over whether dreams have meaning is a contentious topic among psychologists and neurologists and has been for decades. While some experts believe that dreams are simply the result of neurons in the brain firing at random, other researchers believe that dreams hold deep meaning for our lives and are even linked to our overall health. There’s some evidence to back up the argument. Studies have found that dreaming helps us process emotions, turn our short-term memories to long-term ones, and potentially ward off depression.
In the tradition of Sigmund Freud and Carl Jung, some psychologists today still use dreams in therapy to help people better understand themselves and their psyches. Rubin Naiman, PhD, is one such psychologist. A sleep and dream expert with the University of Arizona’s Center for Integrative Medicine, Naiman not only advocates for the power of dreams but also believes society is in the midst of an “epidemic of dream loss,” spurred on by the collective brush-off our culture has given to these nightly visions.
“We are so dismissive about dreams,” he says. “It’s a huge issue. It’s kind of like pretending the moon doesn’t exist.”
In the 1970s, Oldis was involved in early dream research that focused on whether dreams had an effect on our waking lives. It was at this time when researchers discovered that people bodies move while they dream. While this was important information, it became newly relevant to Oldis in 2014, when he read about researchers in Japan who were capturing images from dreams using functional magnetic resonance imaging (fMRI). That got him thinking: Why not also record their movements? And while you’re at it, why not speech as well?
“If you put that all together, what you have is a movie,” Oldis says.
That became his new research question: Can dreams be turned into watchable movies? This led Oldis to partner with the University of Texas at Austin’s Cognitive Neuroscience Lab in 2017 to track the motor behavior of a dreamer. By using an electromyogram (EGM) to measure nerve impulses to the muscles, the researchers successfully recorded the movements of a subject’s dreams, which Oldis turned into a walking avatar that he presented at the International Association for the Study of Dreams in 2017?—?the first step to making a dream movie.
“This is like the early years of the space race,” he says. “But in this case, we’re going into the dream space.”
Oldis has also done some early work around recording dream speech. By attaching electrodes to a sleeper’s speech muscles, he was able to pick up some elements of their dream dialogues, though full sentences and meanings could not be deciphered. He plans to run another experiment at the University of Texas at Austin this fall that he hopes will result in complete speech readings.
In addition to conducting those experiments, Oldis has compiled a team of nearly two dozen dream researchers, neurobiologists, sleep scientists, and psychologists (which he calls “the dream team”) from academic institutions and organizations across the country?—?including the UCLA School of Medicine, the U.S. Army Research Laboratory, the University of Texas Health Science Center at San Antonio, and more?—?to put all the components together and create the first dream movie. Next summer, the team will take over a recording studio in Burbank, California. They’ll bring in a mobile MRI and attempt to record the movement, speech, and images from the full dreams of three or four sleeping subjects.
It’s unclear as of now whether this will be possible, and there’s a lot of work left to be done before it is. The biggest challenge Oldis faces is recording images. Though he hasn’t run an experiment to capture what the dreamer is seeing, a team at Kyoto University has. Led by neuroscientist Yukiyasu Kamitani, the researchers found that through the use of fMRI, “deep images” can be reconstructed from a waking person’s mind?—?when they think of an owl, for example, the vague shape of an owl can be seen.
The way it works is that the team built a computer proxy for the brain, a deep neural network (DNN) that can pull information from different levels of the brain’s visual system, from the simple visual of light all the way to more complex visuals like faces. From there, the DNN essentially reconstructs these layers to recreate images from the brain. Eventually, Kamitani wants to use his research to record the images of dreams.
Oldis’ dream team is not working with Kamitani, but next summer, they plan to use a similar method to reconstruct the visions of a lucid dream, or one in which the dreamer is in control and aware that they are dreaming. This will allow Oldis to tell the dreamer what to dream and test whether the movement, speech, and images he manages to record match up with the actual dream.
Of course, everything Oldis has done and plans to do thus far is a test to see if it’s possible to record our dreams. The other piece of the puzzle is why?—?and what might be gained from this kind of technology?
“Like the moon, it’s there. So you gotta try,” he says. “As Kennedy would say, ‘We’re going to get to the dream in this decade not because it’s easy, but because it’s hard.’ And hopefully it’s of value.”
Not everyone thinks it is. Naiman has been recruited to join Oldis’ dream team in order to play what Oldis calls the devil’s advocate role—or, as Naiman puts it, the angel’s advocate.
“There’s something demonic in what he’s doing,” Naiman says. “I don’t mean that literally?—?I respect what he’s doing… But the downside to it is there are so many attempts to represent the dream in waking life rather than to enter the dream directly. The way we approach dreaming is we pull the fish out of water. But eventually we want to learn to breathe underwater, don’t we?”
At the same time, Naiman laments the sidelining of dreams in our culture and says anything that brings attention to them could be useful, a sentiment echoed by fellow dream researcher and psychologist Deirdre Barrett, PhD, of Harvard Medical School.
“The benefit I see is a new way of seeing dreams,” she says. “And seeing dreams as some piece of our objective brain content could add something to the information we have on them.”
If Oldis’ vision of accurately recreating a full dream as a digital recording comes to fruition, Barrett also says there could be some therapeutic function. Because some people don’t remember their dreams or remember only bits and pieces, being able to play them back in full could add important insight into a patient’s state of mind. But, she says, this is all hugely speculative. After all, the recordings would have to be extremely accurate to capture more than just the basic movement and speech of a dream, and psychologists who work with dreams are more focused on the emotional content than the physical.
“None of these groups are trying to make emotional data a part of it, and there’s not an obvious visual or auditory representation of that,” Barrett says. “And yet, as we know from our own dreams, what we’re feeling in them is a hugely important part. The emotions are a lot of what we’re talking about when we’re talking about our dreams.”
Oldis’ project is still at the beginning stages. It could be 10 to 20 years before there’s technology that can record a full dream movie, he says, let alone a way to capture dreams from the comfort of people’s homes. Until then, it’s unknown what dreams may come?—?or what scientists might do with them when they do.”