Engineering Dreams: Should We?
I’m just back from MIT’s Dream Engineering Symposium, and I only noticed my conference badge for the first time as I was packing to leave. It reads: “Leslie Ellis, Independent.”
I love that. Everyone else had impressive institutional affiliations – MIT, Stanford, Harvard, sleep labs and research centers. Here I am, someone who spent 25+ years in clinical practice working with people’s actual dreams, who now teaches and writes about dream and nightmare treatment for the benefit of other clinicians. I was one of the few who isn’t primarily a researcher.
Still, my talk was very well received. They liked the phrase I used: “the radical intelligence of dreaming.” It seemed to land in a room full of scientists discussing how to capture, guide, and “engineer” dreams to make them more productive, more lucid, more controlled. I suggested we might want to collaborate with dreams, not control them.
Let me show you what I mean.
I told the story of Anya, a refugee who experienced a recurring nightmare of being trapped in her home as the walls crumbled and started caving in. There was no means of escape. But when she was invited to re-enter the dream slowly, with support and calm curiosity, something shifted. She noticed, for the first time, an open window high up, with fresh air, sunlight and birdsong streaming in. It was transformational moment that changed the dream from that day forward.
In my talk, I used an analogy from the equestrian world to explain this approach. There’s a trend called Liberty where instead of breaking a horse to your will with bits and spurs, you let it choose how to respond to your presence. You create a partnership. Dreams are like horses: more powerful than us, with their own desires, wildness and intelligence, but deeply responsive to how we meet them.
Here’s the thing: I didn’t suggest that window. The dream did. This is what I mean by collaboration rather than control, and it’s what I’ve witnessed again and again over decades of clinical practice. When we create the conditions of safety and presence, dreams respond. They change themselves.
The question of control
The Dream Engineering Symposium (DxE) held at MIT Jan 9-10 brought together researchers, clinicians, artists, engineers and sleep specialists to explore a provocative question: can we control our dreams? More to the point, should we?
Dream engineers are already doing it – using a combination of technology and timely prompts to boost creativity, reduce nightmares and increase the chances we can become lucidly aware when we are dreaming. The benefits are real and undeniable: reduced nightmares, improved sleep, increased well-being and cognitive flexibility. Some of these gains come directly from increased agency over our sleeping lives.
Here are a couple of examples, from our nightmare panel. Garret Yount, a scientist at IONS spoke about his study of an online lucid dreaming workshop which significantly reduced nightmare symptoms and distress for a group with PTSD. Westley Youngren, assistant professor at UMKC spoke about how targeted dream incubation can improve dream self-efficacy.
But there’s a shadow side to these advances. What if dreams serve an important biological function that is best not disrupted? In a world where there is already constant competition for our attention, are we trying to harness the one last bastion of peace and quiet to the yoke of profit and productivity?
Michael Grandner, director of sleep programs at the University of Arizona, put it bluntly: sleep is not a physiological process but a non-negotiable biological state for the maintenance of human life. “We don’t sleep and dream because we like it but because we don’t have a choice, and it exists for a reason. Americans devalue sleep because it’s viewed as unproductive time. I think we should be cautious of squeezing value out of it.”
This tension between harnessing and honoring ran through the entire symposium.
The radical intelligence of dreaming
I have spent decades carefully tending dreams and am often stunned by their creative genius. They artfully weave our most salient emotions and current concerns with influences from our bodily state, our sleep environment… and they pull elements from a vast swath of memory, well beyond what we can consciously access. Dreams weave all of these strands into densely meaningful vignettes rich in immersive detail. What they routinely achieve is stunning. How and why they do it is not fully understood.
That’s partly why I advocated for collaboration with dreams, not control. Even in the lucid state, where real-time dream control is possible, long-term lucid dreamers often prefer to let the dream itself guide them. I spoke to one frequent lucid dreamer who said he found his ongoing intention-driven forays into lucid dream activities exhausting and ultimately felt much more settled in dream and waking life when he learned to go with the flow.
In my version of working with nightmares, we are not so much rescripting the dream as re-experiencing it under different conditions – with more support, conscious emotion processing, and softening. The dream itself responds in real time. It spontaneously becomes a different dream. Like Anya’s window appearing where there had been only crumbling walls.
There are real benefits from respectful attempts to work with dreams. In Tibetan dream yoga, lucidity is the first in a series of dream tasks that include passing through solid objects, flying, conjuring and meditation. The result is an increase in cognitive flexibility, precisely what the dream engineers are after, but achieved through cooperation rather than engineering.
New ways to engage with dreams
Many of the experiments presented at DxE were aimed at helping dreamers achieve a lucid state, in part to help the dreamer (perhaps change their nightmares from the inside), and in part to open a real-time window into dreaming and its enduring mysteries. For some, lucidity is a frequent, spontaneous occurrence. For most, however, it takes genuine, sustained effort, and many of the activities that lead to lucidity also disrupt sleep. Still, when achieved, it can be helpful in transforming nightmares and bringing joy.
Another approach is Targeted Dream Incubation (TDI), a method promoted by DxE’s main host Adam Haar Horowitz. This elegant method involves giving a simple suggestion to someone at several intervals while they are falling asleep. The benefits are surprising, including a sense of control that can bring joy to those who feel at the mercy of their challenging dreams.
Since we don’t all have access to a sleep lab, Horowitz has developed an app called DUST to offer these sleep onset suggestions along with many other options for improving nocturnal life. The app is now in beta testing for iPhones. I’ve added some tracks meant to be played just prior to sleep to help with nightmares, lucid dream expert Robert Waggoner has added tracks to help achieve lucidity, and more features are being added as the app develops.
Horowitz’s goal is simple: “To change how well you fall asleep and dream. And how joyously you wake up… Dream science can be an everyday tool, like your cell phone. We can use bookends of the night to improve dream and sleep experience.”
When nightmare treatment is medical necessity
For some, improving sleep and dreaming is not a luxury, but a medical necessity. Pouya Azar, an addiction and pain medicine specialist at Vancouver General Hospital, plans to work with Horowitz and the DUST technology to help ease the intense dreaming in early recovery from addiction. Azar started this work during the devastation of the fentanyl crisis – drug overdose is the leading cause of death in British Columbia for those between the ages of 10-59.
There is a strong interplay between sleep and addiction, Azar explained. Craving and withdrawal symptoms are relatively simple to treat, but the psychological elements – the trauma that led to the addiction, and the trauma experiences that occur when living with an addiction – are not so easy to recover from.
Dreams have become a new target of intervention because they are intense triggers in early sobriety. Dreams of using can evoke a strong temptation to relapse. It is also common to experience intense nightmares in early recovery as the body makes up for the REM sleep it missed. REM rebound is temporary, but until the sleep and dream cycle returns to normal, nightmares disrupt sleep and bring a deep sense of despair that makes users want to go back to numbing their pain.
Azar told the story of a patient, Matty, who became addicted to opioids, then fentanyl after these were prescribed for pain. In all of his time spent in recovery, no one had asked Matty about his dreams, although they were a major cause of distress and relapse risk.
“Dreams are never asked about,” said Azar, a 15-year veteran of addiction medicine. He found that simply asking about nightmares was validating for his patients. “I believe we are on the verge of discovering a new modifiable risk factor – our ability to intervene with the dreams.”
Pleasant dreams under anesthesia
One of the more surprising presentations came from Pilleriin Sikka, a post-doctoral scholar at Stanford, who found that a specific kind of anesthesia tends to produce pleasant dreams. People do indeed dream under anesthesia, and they are typically pleasant dreams.
A pilot study of 30 participants showed promise. One participant, whose son had committed suicide by jumping on train tracks, suffered from PTSD and nightmares about the incident. After a brief imagery rehearsal session and imaginal exposure, she was given anesthetic and dreamt that she was walking alongside the tracks with her son, calmly chatting. This soothing, ordinary dream brought relief from her grief and PTSD. Upon follow up, the dreamer felt more hopeful, and her symptoms had dropped to a sub-clinical level.
The technology landscape
The technology to monitor and manipulate sleep and dreaming has advanced dramatically. Wearable devices like headbands, watches, masks and rings are becoming affordable enough that some of the capabilities of the sleep lab are moving into consumer use. This has potential to bring a new wave of research into how to track and intervene in the sleeping mind.
But as Elie Gottlieb noted, “We’ve entered the zeitgeist of the quantified self where we measure everything but change nothing.” The ability to measure only goes so far, and may even create new anxieties. Roughly half of us have trouble getting good, consistent sleep. Gottlieb suggests we need more accurate and nuanced measures, plus assistance from AI, so we can time interventions based on individual needs. “Timing is critical, not just a detail.”
The symposium showcased various technologies – improved sleep earbuds, a face mask that can measure microexpressions while dreaming, devices to capture physiological sleep data. One system in development, called Luciception, will be able to adjust light, sound, temperature, and smell to create more conducive environments for sleep or more immersive dreaming experiences.
My role on the nightmare panel
At the end of the nightmare panel session, someone from the audience asked: what if we could stop all nightmares from happening – should we?
They handed the mic to me. I said this would not be a desirable thing. Nightmares can carry very important messages that something needs our attention. We want them to stop, but not because we artificially block them. Rather because we attend to the underlying concern they are raising in dramatic fashion. They are part of our body’s alarm system.
This is the heart of the difference between engineering dreams and collaborating with them. Dream engineers may be tempted to optimize, to make dreams more productive, more pleasant, more controllable. I want to listen to what they’re already trying to tell us.
Balancing light and dark
As the symposium wound down, we heard from artists about the ethical responsibilities of dream engineering, about the limitations of reducing dreams to narratives only, and about promising new avenues that will allow us to represent, engage with and study dreams in expansive new ways.
A panel on spiritual aspects of dream study brought some provocative ideas. Michelle Carr, Assistant Research Professor at Université de Montréal, presented her study of experienced lucid dreamers who were asked to elicit encounters with a dream guide. Over 4 nights, participants each had 11 dreams on average, 7 lucid and 3 with dream guide experiences. The latter were experienced as more transcendent and positive.
Andrew Holecek, a scholar-practitioner in Tibetan Buddhism and lucid dream expert, closed the panel with a warning. Too much emphasis on the positive in dream engagement might be “an insidious form of light pollution,” he said. With respect to dreams, we need a balance of dark and light. We need to include the feminine, and she is associated with darkness.
“Perhaps the mercurial nature of dreams is their real message,” Holecek said. “Dreams naturally flee from the light – we badger them for their subliminal messages. Don’t let your quest for dream engineering pin down your dream world. Let’s meet and respect the dream on its terms.”
It was the perfect ending to two days of grappling with how much we should try to control our inner lives. The dream engineers have given us remarkable new tools. The question is whether we have the wisdom to use them with the restraint and respect that dreams deserve.
I keep coming back to those horses. Power paired with sensitivity. Intelligence that responds to how we show up. Dreams are not problems to be solved or resources to be mined. They are partners, if we’re willing to meet them on their terms.