Dream hacking has been a staple of science fiction old and new. In Le Guin’s The Lathe of Heaven, the protagonist, George Orr, discovers that his dreams can alter reality; in Gaiman’s Sandman series, Morpheus himself is the main character; the blockbuster movie Inception involves characters traipsing through each other’s dreams; and dream optimization for mathematical breakthroughs is the crux in Ratnakar’s novella “Axiom of Dreams” in Clarkesworld Magazine. Science fiction, of course. Yet, recent work on the neurobiology of dreams and potential dream manipulation hints that the realm of Morpheus may be more traversable than it’s been considered so far.

Before we enter dreamland, a quick primer on sleep. When we (finally, in this screen-riddled world) fall asleep, our brains and the rest of our bodies go through a set of sleep cycles. In humans, each cycle tends to last between seventy and one hundred and ten minutes, with those earlier in the night on the shorter side, and the later ones a bit longer. In infants, sleep cycles are shorter, coming in at fifty to sixty minutes. Each cycle has four stages: non-REM1, non-REM2, non-REM3 (aka slow-wave sleep), and REM. Creative nomenclature, I know. This last stage, REM, or rapid eye movement, is the topic of much research. It is the final part of a sleep cycle, lasts about ten minutes in the first cycle, and gets progressively longer in the following cycles.

REM is physiologically weird. Our brain starts to produce large amounts of the neurotransmitter acetylcholine and holds back on the release of histamine, serotonin, and norepinephrine. The brain stem shows bursts of electrical activity, and our bodies can show large fluctuations in respiration, body temperature, and circulation. We also lose muscle tone, effectively paralyzing us. This is why sleep paralysis, in which a person maintains consciousness but experiences full-body paralysis, is seen as a (likely) dysfunction of REM sleep.

And, of course, we dream—we all do, even if we don’t remember it. The average person has about three to five dreams per night. We can also dream during non-REM sleep, but those dreams appear to be less active and vivid than REM dreams. The amount and duration of dreams tend to decrease with age. Babies tend to spend half of their sleeping time dreaming. In the elderly, that drops to about a fifth of sleeping time, which coincides with a relative decrease in REM sleep.

It’s not just humans either. In addition to ourselves, other land mammals and birds likely experience vivid dreaming during REM sleep. Even the cuttlefish, Sepia officinalis, a close relative to the octopuses, appears to experience something akin to REM sleep. The fact that REM sleep and dreaming appear to be quite widespread among animals with a sufficiently complex neuroarchitecture suggests that it might serve an evolutionary purpose. In biologists’ language, the capacity to dream might well represent an evolutionarily conserved trait.

What could be the evolutionary purpose of dreams, though?

Perchance to (Lucidly) Dream

Our slumbering adventures have long captured scientific imagination, but only since the advent of modern neuroimaging methods have dreams become a tractable research topic. While we have learned much about brain activity during dream states, the neurotransmitters involved, and so on, the adaptive function of dreams (if any) remains a topic of discussion. The three traditional leading proposals are problem-solving, threat simulation, and memory consolidation. Another recent proposal is that dreams are purposeful noise. Dreaming, in this view, prevents over-fitting by introducing noise into our daily patterns of received stimuli to support our abilities of cognitive and perceptual generalization—dreams give us the ability to adapt our behavior to unexpected circumstances.

If any (or several, for they need not be mutually exclusive) of the adaptive functional accounts for dreaming holds, this suggests that dreams have a potential learning component. Problem-solving, threat simulation, memory consolidation, or ensuring generalization abilities all rely on, and contribute to, the capacity to acquire and implement cognitive skills, from manipulating information in our working memories to integrating perceptual and attentive stimuli. While not the focus of this essay, the importance of dreams in indigenous cultures is often tied to seeing dreams as a way to connect with the spiritual world and obtain deep knowledge.

Regardless of which (combination of) functional account(s) of dreaming is correct, they cast us, the dreamers, in a passive role. We do not control our dreams; we are carried along by brain waves.

There is an exception to this passive role of the dreamer: lucid dreaming; that is, being aware that you are dreaming while in a dream state and having some control over the events in the dream world. Such lucid dreams are generally considered to be states of consciousness between waking and sleeping, with increased brain activity in gamma band frequencies—associated with processes involved in consciousness—and an increase in global cognitive networking.

Roughly half of us will never experience a lucid dream, twenty percent of people perhaps once a month, and only a small percentage of surveyed individuals dream lucidly one or more times per week. Similar to regular dreams, traces of lucid dreaming can be found throughout historical records, but it is only relatively recently that they have become tractable for scientific study. The initial evidence for lucid dreams as a valid phenomenon relied on subjects going through specific eye movement patterns during REM sleep. This simple technique has since been used to investigate whether lucid dreaming is an acquirable skill.

A variety of lucid dream induction methods have been tested, but for the majority of these, the evidence is insufficient and ambiguous, except for the MILD (mnemonic induction of lucid dreams) method. (Or wait a few years, and there could be a lucid dreaming device for sale.) MILD is what’s called a prospective memory technique, or: it requires you to remember to do something in the near future. The most common way to practice it is via the three R’s: rescript, rehearse, and remind yourself. In short, when you wake up, regardless of whether it’s in the middle of the night or because of your morning alarm, immediately try to recall your dream. Then, rescript it (as in rewrite the story of your dream). Rehearse the rescript. Finally, before you lay yourself to sleep again, remind yourself that you’ll be dreaming.

There is another question related to learning and lucid dreaming that we can ask. If lucid dreaming involves specific changes in brain activity reminiscent of waking, can we learn inside a lucid dream? If practice and skill refinement are indeed a matter of repetition, can we train ourselves in a chosen skill while lucidly dreaming? While studies on lucid dreaming have to deal with low sample sizes and potential placebo and motivational effects, tentative findings suggest that, for example, motor practice during lucid dreaming can bolster motor skills and—anecdotally—improve athletic performance.

What Dreams May Come?

However, is this bridge between dreams and waking restricted to lucid dreams? After all, most of us are not lucid dreamers.

Many ancient cultures had rituals that involved people trying to “incubate” their dreams so that they could plea to their deities to provide divine revelation during sleep. One of the oldest written accounts of such a presleep ritual can be found in the Chester Beatty Papyrus, which details an ancient Egyptian way to invoke the wisdom of the dwarf deity Bes. Draw Bes on your palm, swaddle that hand in cloth, and the god will visit you in your dream. Similar rituals were common among ancient Greeks and Romans, as well as in many indigenous cultures across the world, notably the Dreamtime journeys in Aboriginal Australian cultures. Often, these rituals also involved sleeping in holy places and so-called cutaneous stimulation, for example by sleeping on covers with an unusual texture or rubbing mud on one’s skin.

We don’t have to stick to divine revelations. How about scientific breakthroughs? German chemist August Kekulé famously solved the structure of the benzene molecule following a dream of ouroboros, the snake that eats its own tail. Without Niels Bohr’s dreams, we may have had to wait longer before unraveling the atomic structure (which landed him the Nobel Prize in Physics). Elias Howe revolutionized the sewing machine after dreaming about cannibals that brandished spears with holes in their tips. Dmitri Mendeleev ordered the entire chemical universe in a periodic table that he dreamed almost fully complete. It’s not only scientific breakthroughs that come in dreams either. Paul McCartney dreamed the beginning of the melodies that would become the monster hits “Yesterday” and “Let It Be.” Another monster hit began with a nightmare by Mary Wollstonecraft Shelley. The result, Frankenstein; or, The Modern Prometheus, is considered by many to be the beginning of modern speculative fiction. Robert Louis Stevenson found the clues to his Strange Case of Dr. Jekyll and Mr. Hyde in his slumber, and, more recently, Stephenie Meyer credits a dream for her hugely successful novel Twilight.

I could go on and list more creative works and scientific insights that existed only in dreams before they became real, but I think this is enough to make my point: dreams and waking life are not separate planes of existence.

Still, all of these examples are the result of coincidence, a haphazard burst of dreamed inspiration. A flurry of research in the past few years, however, shows how technology can (potentially, such is the way of science) improve our ability to pierce the barrier that separates our dreams from our waking lives.

The idea of leveraging technology to induce, incubate, or influence our dreams goes back, though. In the 1960s, artist Brion Gysin teamed up with engineer Ian Sommerville to build the Dreamachine, which used a stroboscopic effect to induce dreamlike states in observers. But it’s only since the early 2010s, with the advent of wearable EEG devices, that our dreams truly came within reach. True to its reputation as a (dream?) incubator, Silicon Valley was the location of the 2015 Consciousness Hacking event with a panel on Dream Tech. The goal? Using the wearable EEG technology to induce lucid dreams through biofeedback. It turns out to be a bit more complicated than that. And yet, EEG and related technologies continue to pull back the veil that separates dreams from reality.

For example, a 2023 study found that it’s possible to “communicate” with dreamers by telling them real or made-up words and asking for a response (frown or smile). And that’s only the beginning. In a case of reality following fiction, another 2023 study found that dream incubation is very real. By presenting auditory cues at sleep onset, the researchers improved creative performance on tasks that assessed semantic distance (a measure for divergent thinking) by the sleepers after their slumber. Add a dream glove that detects when you enter the non-REM1 sleep stage (the first stage of the sleep cycle and a creative sweet spot), and the creativity of your own mind might baffle you. Salvador Dali and Thomas Edison are well-known examples of people who devised ways to rouse themselves the moment they drifted off to leverage the creativity that the non-REM1 stage granted them. You may not even have to wake up to take advantage of this. When machine learning meets MRI scans, it might become possible—in another example of fiction becoming reality—to “read” your dreams.

Finally, and most speculatively, dreams can potentially fight prejudice and improve our empathic abilities. It has been theorized that reading fiction improves empathy. As of yet, there is little work testing this in a dream setting, but a 2019 study does suggest that dreams, through acting as a form of fiction, might improve the capacity to feel empathy. While the researchers found effects on so-called state empathy (empathy for situations rather than for specific personal traits) in individuals recounting their dreams and listening to others doing the same, it is possible to extend this idea and hypothesize that dreams themselves affect the empathy of the dreamer.

Of course, all this research concerns small-scale trials in controlled conditions. It remains to be seen whether or not dream hacking in real-life conditions is feasible or desirable. In a final return to science fiction and fantasy, it’s a small step from hacking to harvesting. In N. K. Jemisin’s Dreamblood duology, people’s dreams are harvested for magic that keeps the city-state Gujaareh running. But if dreams become a resource to exploit, what’s to stop the powerful from enforcing sleep on people? More futuristically, what if the same techniques and technologies that enable dream incubation also allow others to infiltrate your dreams and, who knows, steal proprietary information, Inception-style? This remains, for now, science fiction.

Yet, the possibility of advertisers hacking or incubating our dreams is realistic enough for a group of sleep experts to sign an opinion piece warning against it after a study came out in which a wearable electronic device almost literally planted trees in people’s dreams. Or, because capitalism is going to try to capitalize on everything including dreams, what about a venture-backed startup that wants to keep people working in their (induced) lucid dreams?

Will our dreams provide a wonderful new creative world for us to intentionally explore, or will they become another target to extract corporate profit from? Or both?

Sweet dreams.

Share this page on: