Cobwebs of the Mind

What science can tell us about dreams and consciousness.

Mary Shelley’s Frankenstein has long held pride of place on my bookshelves and in my heart. But with each passing year, the premise of the novel seems to become less fiction and more science — or at least, scientifically plausible.

Plenty has been written about the biology behind the story, about galvanism and anatomical dissections in the nineteenth century, xenotransplantation and electrical stimulation of cell division in the twenty and twenty-first. There is even a YouTube miniseries that modernizes the plot with real scientific advances, about how 3-D organ printing, cell grafts and other wonders could give rise to reanimated resurrections.

Less has been written about the biology behind the story’s origins. About the dream that inspired it all. As Shelley herself wrote:

“When I placed my head on my pillow, I did not sleep, nor could I be said to think. My imagination, unbidden, possessed and guided me, gifting the successive images that arose in my mind with a vividness far beyond the usual bounds of reverie…I saw the pale student of unhallowed arts kneeling beside the thing he had put together. I saw the hideous phantasm of a man stretched out, and then, on the working of some powerful engine, show signs of life, and stir with an uneasy, half vital motion.”

So how does the sleeping brain churn out such strange sights? How is it that dreams can give rise to such horrors while also being synonymous with our greatest hopes and desires?


Enter rapid-eye-movement (REM) sleep.

One of the four currently established stages in the sleep cycle, REM sleep is characterized by some weird biology. The eponymous rapid eye movements are jerky, almost blink-like motions of the eye during certain periods of sleep. These erratic movements are accompanied by fluctuations in pulse, blood pressure, breathing patterns and brain activity that fly counter to our cultural leitmotif of sleep as a sort of death-adjacent state.

In fact, some neurologists have reported that electrical activity in the awake brain can’t be differentiated from electrical activity during REM sleep. Which probably explains why dreams are so immersive, harnessing all five senses and the fullest extent of our emotional responses to construct experiences that feel indistinguishable from waking reality.

But for all their palpable realism, dreams also have a distinct surreality to them. Until science advances significantly, we’re probably not going to encounter English-speaking dinosaurs riding airplanes over oceans of peppermint-marshmallow ice cream anywhere outside the dreamscape.

Robert Stickgold’s NEXTUP model offers a possible explanation, as well as interesting insights into dreams’ effects on memory consolidation and cognitive processing. It seems that REM sleep prioritizes what are called “weak associations” between ideas. For example, the words “right” and “wrong” have a strong association. However, “wrong” and “thief” are more weakly associated. Not in any moralistic sense, but in the sense that the word “thief” is lower down the list of words that come to mind unprompted when we hear the word “wrong” than the diametric opposite “right”.


Scientists still haven’t figured out the association between these movements and dreams. Some hypothesize that our eyes move to track the activity in our dreams, much like how they move in day-to-day life. Others believe the REMs come first, and that dreams are our brains’ attempts to try and make up a visual story to go with them.

That same chicken-egg debate applies to our sense of space and time in dreams. It might not be a coincidence that mechanisms so heavily intertwined with our emotional state and activity levels as our heart and breathing rate also fluctuate during REM sleep, even as we dream of dancing on distant planets populated by floating lights or running from shadowy monsters lurking on the fuzzy edges of our awareness.

While the idea that dreams result from these physiological fluctuations explains why we don’t have a strong record of people smelling or tasting things in dreams, it wouldn’t explain some other important things. Such as the fact that we dream in other, nonREM stages of sleep, too.


Sleep is a cycle. Researchers have found four distinct stages of sleep, each of which are cycled through roughly every 90 minutes. Besides REM, we also have the N1, N2 and N3 stages.

Some researchers hypothesize that the increase in weak association experienced during sleep is due to the fact that serotonin release is almost completely blocked during REM sleep. Which might explain associations between the less dramatic reduction in circulating serotonin levels and the less dramatic dreams that occur during N1 sleep.

But what is serotonin and why does this one molecule matter so much? The short answer is that serotonin is kind of like a neurochemical Swiss Army knife. It can serve a lot of different purposes. For example, shortages of serotonin have been implicated in everything from major depressive disorder to hallucinogenic drug-induced “trips”.

Enter N1 dreaming. Serotonin levels drop during N1 sleep as well, suggesting that N1-stage dreaming is predicated on sleep-induced serotonin shortfalls as well. But unlike in the rest of the stages, N1 dreams tend to lack narratives and representations of the self. They are usually made up of geometric patterns, simple pictures or bizarre thoughts without any greater story tying them together.

Sound familiar?


But you don’t need drugs to enter the dreamscape outside of sleeping hours. You don’t even need to close your eyes.

See, REM sleep is functionally and neurologically indistinguishable from daydreaming. Domhoff and Fox found that the parts of the brain activated during dreaming, called Default Mode Networks, are the same ones activated when you zone out during boring lectures and meetings. These DMNs are demonstrably linked to introspection, aesthetic appreciation, social development and have been repeatedly shown to have strong connections to emotional processing.

Which would certainly explain how nightmares seem to be such perfectly calibrated collages of all our most primal fears and anxieties.

(It would also explain why “emotional salience” is an understatement for the visceral revulsion that even the merest mention of Danny Boyle’s iconic 28 Days Later invokes in me to this day.)

Interestingly, DMNs are also implicated in both recalling past memories and imagining future situations. There is evidence that they act as a sort of rehearsal mechanism, making sure that we analyze and (hopefully) don’t repeat historical mistakes.

It seems an idle mind is really a problem-solving workshop. Much like Santa’s elves , our Default Mode Networks create whimsy and wonder, refitting and adjusting weak associations and processing information to come up with new ideas and approaches. Between NEXTUP and DMN research, it seems that dream states are linked to real cognitive benefits and innovation.

Something to keep in mind next time your professor or boss berates you for not paying attention.


There are some questions to be asked about the importance of narratives, though. Is it true that most of our REM dreams have narratives, or is it simply that we remember the dreams with narratives more? There is plenty of psychological research that demonstrates how creating stories with plots makes it easier for people to remember lists of words or scientific concepts (a process known as “narrative chaining”). What is to say that the same doesn’t hold true for dream recall?

In fact, Shelley describes her state as not asleep but also not thinking, a description reminiscent of the amorphousness of N1 or hypnagogic “dreams”. As Stickgold and Antonio Zadra describe it, N1 dreams are “often clearly related to the thoughts you were having immediately before falling asleep”. It serves as a seemingly seamless transition stage, assimilating our conscious, voluntary thoughts into a temporal limbo between waking reality and the dreamscape.

While many people know the old authorial spiel of “it came to me in a dream”, they rarely recognize the synapses and associations that had to be formed prior to the dreaming. While Mary Shelley’s iconic novel was inspired by a dream, that dream itself was inspired by

“the conversations between Lord Byron and Shelley, to which I was a devout but nearly silent listener. During one of these, various philosophical doctrines were discussed, and among others the nature of the principle of life, and whether there was any probability of its ever being discovered and communicated…Perhaps a corpse would be re-animated; galvanism had given token of such things: perhaps the component parts of a creature might be manufactured, brought together, and endued with vital warmth.”

The association between dreams and creativity are tied to these functions, which allow us to tap into possibly unexplored weak associations that can go on to inspire works of visual art, music and literature. But it’s rare that a narrative comes fully formed. The skill of the creator lies in transforming and refining those disjointed narratives and dream experiences into a compelling work.


Of course, dreams often elude our own attempts at remembering them using such limited things as language and sensory descriptors. Which is why Zadra and Stickgold emphasize that what we think of as our dreams are not only the series of experiences we have while asleep, but also what we remember and report about these experiences upon waking.

One of the earliest studies of REM sleep found that participants who were woken up in the middle of REM sleep provided detailed descriptions of their dreams 80 percent of the time. Meanwhile, participants woken up from nonREM sleep only managed to recall and describe their dreams only 7 percent of the time. Later research has pushed that number higher, though, with some researchers reporting up to 70% recall for N2 stage dreams.

Those numbers have gone even higher for N1 stage “hypnogogic” dreams, with participants reporting dreams for around 75% of awakenings. That is pretty close to the nearly 80% of REM sleep! But the fact remains that REM sleep (and post-REM recall) is what most science focuses on.

This means that it is important to avoid confusing the forest for a lone tree. As tempting as it can be to treat REM sleep as fungible with dreaming, they are very distinct entities.

REM sleep (as well as the other stages, to varying extents) is physiological, quantifiable, observable. It has clear biological correlates, can be mathematically measured, and easily seen by external viewers. The same goes for post-REM recall.

But the truth of dreams as experiences remains elusive. They are highly subjective, entirely internal and opaque to others. We can only record what others’ report, what we ourselves remember upon waking. But those few images and sensations are rarely the whole story, or even the whole picture.

For now, at least.


Cutting-edge research led by Tomoyasu Horikawa might mark the nascence of dream-watching. Through arduous cataloging efforts and multi-voxel pattern analysis, the Japanese team was able to calculate fMRI signals that correlated to thousands of pictures. Dreaming subjects’ fMRI activity was measured right before waking, and the researchers found that the signals correlated strongly with the images people reported having seen in their dreams.

It makes sense, given that dreaming activity is similar to waking brain activity, that the same parts of our brain that activate while looking at flowers or friends will reactivate to create dreamscape versions of those entities.

This research also has real-life clinical applications that push the boundaries of consciousness as we know it. Adrian Owen’s groundbreaking research on people believed to be in vegetative states revealed that, when asked to visualize things like “playing tennis” or “walking from room to room of your house”, some of the comatose patients show activity in the same brain regions as people who can walk, talk and do the other things we associate with consciousness, who themselves all showed consistent patterns of activation.

Owens’ describes how this research could provide a window into the minds of people society has written off or dismissed due to limited verbal and/or motor function. Stickgold and Zadra detail how such research could be used to better harness dreams for therapeutic ends, a way for scientists and clinicians to understand how the brain forms traumatic connections and associations.

Of course, for all the fascinating new knowledge, there might also be fears that this knowledge is the harbinger of some nightmarish Wachowski-esque dystopia where malevolent powers can read and alter our dreams.

Imagine that, a world where regulated pulses of electrical activity designed to selectively strengthen synaptic connections keep our minds trapped in meticulously crafted dreamscapes. Where a unified set of cultural signifiers and symbols reduces the breathtaking diversity of human imagination to a regimented set of synchronized signals that repeat on loop, locking us into a carousel of prefabricated experiences, on and on, ad nauseum…

Still, I would advise concerned readers to hold off on buying tinfoil hats for a little longer. Real-life dream infiltration a la Inception and Paprika is still far, far into the future. The research outlined above is but the first tentative steps of a growing field flush with wonderful therapeutic possibilities, with information that can help us better understand the limits and expanses of human consciousness.

So until the day when a couple of mad computer scientists figure out how to upload our consciousnesses into the cloud, let’s keep on dreaming.