Targeted dream incubation (TDI) is a methodology for guiding (or “incubating”) dreams towards specific themes. Please read the FAQ below to learn more about TDI.
We engineered a device to carry out TDI called Dormio. We recently published a study using TDI to understand the link between dreaming and creativity.
Targeted dream incubation (TDI) is a method for guiding (or “incubating”) dreams towards specific themes. In our lab, we have mainly explored using auditory and olfactory stimuli to achieve TDI, but the TDI methodology encompasses a wide range of possible interventions to achieve guiding dreams. The TDI protocol also includes serial awakenings that enable collection of dream reports following the incubation of a dream.
So far, we have applied TDI to Stage 1 (also called NREM1 or N1) sleep, which is the first stage of sleep. During sleep onset, hypnagogia (a special state occurring in the transition from wakefulness to sleep) occurs. Sleep onset is characterized by a gradual, piece-by-piece descent into sleep, as opposed to a sudden, binary on/off switch from wakefulness to sleep which some people imagine occurs. In fact, there are nine separate substages of sleep onset. In the middle of this gradual transition from wakefulness to sleep, the brain maintains sensitivity to outside stimuli (for example, people can still hear sounds and smell scents in the space around them) even as the brain enters a more dream-like state, both in terms of physiology and mental experience.
In our work, we use a TDI protocol with auditory stimulation to guide sleep onset (N1) dreams. We carry out our TDI protocol with the Dormio system (read more on Dormio here), but TDI can be carried out with any combination of tools that can track sleep stages, play stimuli, and record dream reports. A person wears Dormio and lies down to fall asleep. The Dormio system tracks sleep onset. Once sleep onset is detected, a timer of a few minutes is started. At the end of the timer, an audio recording is played to ask the user for a dream report, bringing the wearer back into wakefulness briefly. We record everything the user says during their dream report, to avoid them forgetting a potentially useful idea. Following their dream report, the system then plays an audio cue, reminding the wearer to think of certain words (like "fork" or "rabbit"), with the aim of integrating the cued topic into their next set of dreams. The user then drifts back to sleep, with the cue in mind. In our laboratory testing, we have found that the cued words reliably entered the hypnagogic dreams of our users. The system continues to track the state (awake or asleep) of the user, repeating the process described above of waking them up after a few minutes of sleep to collect a dream report. This protocol is carried out repeatedly to guide dreams and collect dream reports.
To better understand how our auditory stimulus makes its way into people’s dreams, just consider “a lion playing volleyball underwater.” This phrase, even written, conjures a mental image. The creation of a mental image from words also happens at sleep onset. Words heard by the napper just as they fall asleep serve as a seed for mental images/thoughts, allowing the ideas to slip into their dream. We wake the subject up after a few minutes and request a dream report in order to avoid them slipping into a deeper sleep, at which point the likelihood they forget their dream would increase.
We began this work using the Dormio device for tracking sleep so we could incubate dreams. Since then we have also used the Masca, the Hypnodyne, and even typical polysomnography to enact TDI and produce targeted dreams. We emphasize that our TDI protocol can be carried out using a variety of technologies to track sleep, play, and record audio. We are working on new ways to do low-tech dream incubation, such as through just an online timer interface: https://christinatchen.github.io/dormio/timer.
TDI is similar to a technique called Targeted Memory Reactivation (TMR). In this TMR, sensory cues are paired with a task while subjects are awake. Then, during a subsequent sleep (which typically includes later phases of sleep such as N2, N3, and REM), these sensory cues are presented again, with the goal of reactivating the memory of the learned material from the task associated with the cue. When the task is tested after sleep, memory performance for items that were cued during sleep is typically better as compared to memory for items not cued during sleep (suggesting important sleep-dependent processing for learning and showing that TMR can affect such processing). TMR is a powerful technique, but it is not focused on changing dreams, and does not involve collecting dream reports. Instead, it is focused on affecting the topics the sleeping brain is working on consolidating, using cues to direct the brain to augment specific memories over others, without regard for how this is experienced by the sleeper.
The first reason is to facilitate personal introspection. We find the idea that there is a state of mind (sleep) which composes and constructs the conscious self, but remains inaccessible to it during the day, both frustrating and alluring. Hypnagogia is a version of oneself that the waking self is unfamiliar with, a self that slips past memory as we drift into unconsciousness. Good neuroscience can aspire to be effective self-examination. Good technology in service of making neuroscience relevant outside the laboratory, then, should facilitate self-examination. The ends of this project are both practical and philosophical. We have no doubt that hypnagogia holds applications for augmenting memory, learning, and creativity. Yet also, after having explored the state ourselves, it seems a deeply valuable and inspiring sort of self-seeing which was inaccessible to us previously. As Nobel Prize winner Eric Kandel said, "human creativity...stems from access to underlying, unconscious forces." To know ourselves, and to be our most creative selves, we are interested in building tools for self-exploration in this sleep state. TDI aims to be a tool for people to use on their own terms to explore and augment themselves.
Beyond personal introspection, the second reason this method is so exciting to us is that it opens up new avenues for exploring the mind. Scientifically, having a method to guide dreams means that we can now do controlled experiments on how dreams and dream content influence cognition, including questions about emotion, creativity, memory, and more. We know from correlational studies and anecdotal reports that different dreams are linked to different post-sleep outcomes – such as creative performance and emotion – but without controlled studies, we still lack solid scientific evidence for a causal effect of guiding dreams to improve these outcomes. So long as we cannot guide dreams, we cannot do controlled experiments on dream content. TDI represents a breakthrough in this long standing methodological hurdle.
A third reason is to open possibilities in the therapeutic domain. TDI may give patients and clinicians a lever of control to gain insight via dreams and to combat nightmares, which take a huge toll on people who struggle with anxiety and trauma. We have already begun a study in collaboration with Westley Youngren and the Veterans Affairs Office to test this application of TDI. Creatively, the rich history of luminaries using their dreams, and specifically the hypnagogic state (Sylvia Plath, Salvador Dalí, Edgar Allen Poe to name a few), to release creative potential points to the possibility of using TDI for targeted creative brainstorming. We've already run one experiment showing TDI can enhance creativity, but the real test is putting it in the hands of creatives all over.
TDI was published in a 2020 paper in the journal Consciousness and Cognition, which was a collaboration between MIT, Harvard, and Boston College. This paper forms part of the journal’s larger Special Issue on Dream Engineering, collecting papers from sleep scientists around the world on their methods for researching and guiding dreams. This Special Issue came out of the Dream Engineering workshop which we hosted at the MIT Media Lab in 2019, gathering a community of researchers from around the world to empower the dream engineering field and link technologists with scientists for new collaboration. The community of dream engineers is growing; a recent paper from the Gori lab outlines a vision for engineering dreams to understand blindness. Together we’re hoping to generate and answer questions about the parts of our minds which can be hard to see, and make the tools that make answering them possible, and build dream-based therapies and interventions. You can read about our vision for the dream engineering movement here.
In addition, TDI was central to Adam Haar Horowitz’s Master's thesis and PhD thesis. The experiments in Adam’s thesis focused on augmenting creativity with TDI (now published in Scientific Reports), using TDI to incubate REM dreams (not yet published), and using TDI to decrease feelings of helplessness/loss of agency related to nightmares (not yet published).
If you want to see what the press is saying about those experiments, here is an article in The Scientist and here is one in MIT News. In other labs, TDI is being used in an experiment on PTSD related nightmares at University of Kansas and in an experiment on mind-wandering and dreams at Duke.
The goal of TDI is both magnetic and unlikely: Can we really “engineer” dreams, our internal worlds that feel so fundamentally out of our control? We are far from the first to be excited about dreams and the possibility of guiding them. The incubation of specific dream content has fascinated people for millennia, from Ancient Egyptian spiritual practices and Canadian Indigenous dream sharing rituals to contemporary treatments for PTSD-related nightmares. While reliable techniques for dream incubation have proven elusive in the laboratory, yet it's crucial to know that targeted dream incubation (TDI) is a modern instantiation of an ancient technique. Here is a great summary of the myriad ways people have historically interfaced with dreams. Understanding that dream incubation is a practice that ranges from sacred to scientific space across cultures, and that it is a potentially powerful way to influence the mind, we are committed to (and continually updating) our ethics on the use of dream influencing technologies.
Dreams are a vast topic, touching everything from consciousness studies to sensor technologies to indigenous healing practices. Here we list a completely non-exhaustive set of potential ways to learn more about dreams, including many scholars who informed and inspired our own studies. Scientists like Stephen LaBerge and Benjamin Baird do wonderful work on later-stage lucid dreaming, focusing on the REM state. Scientists like Jonathan Smallwood, Paul Seli and Jonathan Schooler have done work on mind-wandering and creativity, inspiring our idea that fluid thinking outside of executive control in hypnagogia (like mind-wandering) could augment creativity. Work by Deirdre Barrett compiling moments of inspiration found in sleep, and work by Robert Stickgold and Tore Nielsen on microdream phenomenology, all encouraged and informed us. Andreas Mavromatis wrote a whole thesis on hypnagogia, and his writing gave us a sense of the poetry and practical applications of this state (as did Nabokov, Oliver Sacks, Yoga Nidra practitioners, and Edgar Allen Poe writing on hypnagogia). We also were inspired by Matthew Spellberg’s writing on dream sharing rituals, Kelly Bulkeley’s work on dreaming in world religions, and Robert Stickgold and Antonio Zadra’s book When Brains Dream.