You sleep every day. It is a period of low activity in one spot. You might roll over, but usually you are not walking around. You lie down typically; that is, you adopt a particular posture. You can wake up from sleep, although you are less responsive to disturbance than you would be when alert. If you don’t get your sleep, you will be sleepy the next day and may have an urge to take a nap or go to bed early.
All animals sleep—at least, we have yet to discover cases of an animal that has no cycle of sleeplike behavior. But what is sleep? The actions just noted comprise the definition of behavioral sleep. They show how we recognize sleep in other animals. It is not identical to resting. Being less responsive distinguishes sleep from simply “resting your eyes” or otherwise being stationary.
That you can be woken from sleep distinguishes it from other forms of unconsciousness, like being knocked out or in a coma. Sleep, unlike mere quiescence, is also regulated—you cannot skip sleep without paying the price, and you cannot skip it entirely for long. A certain amount of sleep is required for you to function.
Every species has a typical sleep posture, although these are not the same for each. Sperm whales, for example, sleep suspended vertically head up below the water surface, rising at intervals to breathe. Parrotfish sleep in shelter, ensconced in a cocoon of their own mucus. The upside-down jelly, Cassiopea, lives mouth up on the bottom of shallow seas.
The jellies contain algae cells photosynthesizing in their tissue, from which they derive nutrition. They slowly pulse their bells in the sunshine to ventilate the algae. These jellies sleep in this upside-down position at night when their ventilation slows. Wake them, and the ventilations increase again. If kept up all night by inquisitive scientists, the upside-down jellies fall asleep more during the next day.
That octopuses sleep has long seemed the case. The demonstration of this in the scientific literature is recent though, and studies of sleep among cephalopods are rapidly increasing what we know. The first study to identify sleep in cephalopods was on cuttlefish, and reported only in abstracts beginning in 2002. Another ten years passed before a full paper appeared in 2012, showing that cuttlefish (Sepia officinalis) do sleep. Octopuses got attention in the same timeframe, with important studies in 2006 and 2011.
All animals likely have two phases of sleep. Consider the two stages of human sleep. First, as we fall asleep we enter a liminal state between relaxation and sleep. This lasts just a few minutes as we relax. We settle then into light sleep, and eventually deep sleep with very relaxed muscles. Body temperature drops. Our breathing and heart rate are at their lowest levels and our eyes are not moving under their lids.
Do animals dream? And if so, can we know anything about it?At the next stage, while our big muscles remain deeply relaxed, muscles in the eyes, face, fingers, and toes twitch more. In particular, our eyes move rapidly, as though looking at something. Our heart rate increases but body temperature does not. This is rapid eye movement (REM) sleep. The twitching and eye movements make it a high-activity sleep. The early stages of falling asleep, light sleep, and deep sleep are collectively non-REM, low-activity phases. We spend up to eighty percent of our time asleep in the low activity non-REM phases, and the rest in high-activity REM sleep. We alternate between the two throughout the night.
Low- and high-activity sleep stages also occur in other animals, likely all other animals.
Researchers, inspired by the changing body pattern activity of Heidi when asleep, filmed medium-sized Octopus insularis—a species common in coastal Brazil and its oceanic islands—to learn more about octopus sleep. Often these animals were active or were alert but not active. Sometimes they were quiet but with the pupils still open—not clearly alert but not asleep, just resting.
The appearance of these octopuses was different when asleep. The pupils were closed. Skin color was uniformly pale. The researchers called this quiet sleep, a low-activity sleep phase. Short bouts of active sleep interrupted quiet sleep. The periods of active sleep lasted less than a minute each, compared to quiet sleep bouts that averaged almost seven minutes. During active sleep, the octopus displayed changing patterns of skin color and texture, as well as rapid eye movements just as people exhibit during REM sleep.
This was how Heidi was displaying when asleep. Stationary and mostly relaxed, pupils closed, but arm tips twitching and body patterns changing color and texture. She was asleep, in the active sleep phase. Animal active sleep is not identical to human REM sleep, but there are parallels and similarities.
Often when awakened from sleep, we can relate a dream experience we were having. This happens about half of the time when we awake from non-REM sleep, and about eighty percent of the time when we awaken from REM sleep. Dreaming is a common human experience, and it’s more common during REM sleep than non-REM sleep.
Do animals dream? And if so, can we know anything about it? Or does all knowledge of dreaming necessarily come from people telling us their dreams, so that nothing can ever be learned about animals dreaming until, like Doctor Dolittle, we can talk to the animals?
Understanding whether animals dream is important for at least two reasons. First, dreaming is a form of consciousness. If animals dream, that says something about their being conscious. Second, dreaming is imaginative. In a dream, the experiences of the dream are not related to the present environment. They are constructed—imagined—from past memories and events. An animal that dreams is one that can imagine, which is a capacity that underlies planning and creativity.
Most of what we know about human dreaming come from our own verbal dream reports, but dream researchers use other tools, as well. Telling about your dream opens the door to knowing about it, but not everything that we know about dreaming comes from telling the dream.
A dream is a mental experience that occurs to the dreamer during sleep. It is subjective—only the dreamer has direct access to the dream. Dreaming is not cut-and-dried distinct from other kinds of subjective experiences such as daydreaming or mind-wandering or psychedelic experience.
Odd contradictions would be inherent if dreaming is defined only in relation to telling someone about it. If I tell no one of my dream, would that mean it didn’t happen?Odd contradictions would be inherent if dreaming is defined only in relation to telling someone about it. If I tell no one of my dream, would that mean it didn’t happen? Conversely, if I lie (even to myself) that I had a dream I did not experience, would that mean I really did have the dream? If I have a dream that affects my mood, but forget the dream before I can even tell it to anyone, my altered mood persists.
The dream cannot depend on the telling. Rather, the reverse is true. Indeed, the experience in a dream and the memory of the experience depend on different parts of the brain. That suggests we can know something about dreams without the telling of them, even if the dream is forgotten. This indeed turns out to be the case.
There are at least three ways that we learn about human dreaming besides the telling of the dream. We act out some dreams in our sleep. Our brain activity changes. And dreaming helps us learn.
First, sometimes the normal inhibition of large-muscle movement during sleep breaks down, and people, for example, sleepwalk. In this state, asleep but not inactive, sleeping dreamers may act out their dreams. Such enactments are reported by the vast majority of people asked. In such cases, violent sleep behaviors accompany violent dreams. Postpartum women may dream of looking for their baby, and feel around the bed in their sleep. Those who talk in their sleep, if awoken, often describe a dream related to their speech.
These sleep behaviors thus sometimes reflect dream experiences. This is true whether the dreamer can tell of the dream or not. Dreams often are hard to recall, in part because dream experience and dream recall can function independently of each other. One symptom of a sleep disorder in people is the increased muscle activity that allows dreams to be acted out.
Patients with REM sleep behavior disorder may act out dreams that they don’t recall. That is, patients will enact a dream while asleep, but do not recall a dream on waking. It is reasonable to conclude these patients were still experiencing the dream during their scenic movements while asleep, but did not retain a memory of the dream.
The second way to learn about dreams without telling them is to look at brain activity. There are patterns of activity in the brain that accompany dreaming, associated both with a specific brain region and with particular patterns of neural activity. Dreaming stops in patients with damage to this area of the brain, but can be unaffected by damage to other brain areas, even if those damaged areas otherwise affect sleep.
Dream experience is predictable from activity in the colorfully named posterior hot zone of the brain. This strong connection reveals that we often have dreams but don’t recall them, as those two functions require activity in different parts of the brain.
Brain region activation also gives some hint of dream content. Activity occurs in the fusiform face area of the brain when dreaming of faces, in the brain area processing spatial relationships when dreaming of particular spatial settings, and in the areas of the brain involved in perceiving motion when dreaming of movement. In principle, detailed scanning of brain activity could reveal when a dream is experienced, something about its content, and whether it might be recalled on awakening. This level of knowledge from brain scanning is difficult to achieve in practice.
The same patterns that fire in the birdsong system of the brain while the finches are rehearsing aloud also replay at the same tempo in the brain while the birds are asleep.Our brains reactivate recent memories during sleep, a process called replay. During the original awake experience, a particular pattern of neural activity occurs, forming a new memory. During sleep, the same neural patterns may be replayed, and will unfold at the same pace as the original experience. This occurs in humans, but has been more extensively studied in birds and rats.
People learning new motor skills who also sleepwalk will do the same thing behaviorally—partially reenacting the new skill while sleepwalking. We further reexperience our newly learned memories in dreams. Thus newly formed memories are replayed three ways during sleep that parallel each other: replayed brain activation patterns, replayed sleep behaviors, and replayed experiences in dreams.
The third way to learn about dreams without telling them arises from the effects on learning of the replay of neural patterns in sleep. Zebra Finch songs are not innate; the finches must learn their songs from other Zebra Finches. To do this, they rehearse aloud, improving their imitation as they practice. The same patterns that fire in the birdsong system of the brain while the finches are rehearsing aloud, researchers discovered, also replay at the same tempo in the brain while the birds are asleep.
The sleeping finches further moved their vocal cords as they had during song production but silently. Finally, the auditory center of the brain responded as though hearing the soundless neural pattern. The birds were silently rehearsing in their sleep, replaying and also “hearing” the same patterns that produce song when awake.
Birds sometimes do vocalize in their sleep. Sleep-singing has been noted since Roman times. If a bird could talk, might we know what they were dreaming? Alex the Grey Parrot learned an extensive use of words in the interactive setting of a long-running behavioral study conducted by Irene Pepperberg. Alex could talk. He understood words in context. He vocalized. He used words to ask for his favorite treats or toys.
During attacks on prey, the octopus employs a series of body patterns, different from the patterns seen when a predator attacks an awake octopus.Alex rehearsed his new conversational skills in monologues, whether or not there was anyone listening, much as children do when they are learning language, and much as the Zebra Finches do when learning songs. His vocalizations were recorded at the start and end of the day when he was alone, times in which he did a lot of rehearsing. There is no video from these recordings, so we do not know that Alex was always awake while he practiced. From a talking bird, trained like the sleep-rehearsing Alex, we might learn whether birds are dreaming of more than their song and their learning.
*
What of Heidi asleep in my living room, arm tips twitching and body pattern displays moving across her body in waves and sudden starts? We are not certain whether she was dreaming, or of what . . . but could we know this? She cannot tell us verbally of her dreams. Obtaining brain activity information from octopuses in salt water is difficult—such data have been reported from only a single octopus.
Body pattern changes are a form of sleep behavior, however. Much like talking in our sleep or sleep-singing, body pattern changes do not involve the deeply relaxed large body muscles that are inhibited during sleep, which usually prevents whole body motion. Octopus body patterns are muscularly controlled. These muscles are small, like those controlling our own faces, fingers, and toes that twitch in REM sleep. Teaching Heidi a new behavior associated with a body pattern change might lead her to replay that pattern in her sleep, revealing dream content. But already, her body pattern may be enacting her dream.
Octopus body pattern displays occur in particular circumstances: octopus displays encountering a predator are different to that when pursuing prey or finding a mate or exploring the reef. Earlier I described the body patterns of an octopus while escaping an attack by a moray eel—much of it blanched white, turning dark in flight. An octopus attacking prey may move from camouflage to passing cloud to web blanched (but not the mantle) to camouflage again. The detailed sequences of these waking patterns are not well-studied. Heidi was not able to tell me of her dream, but, possibly, she was able to show me.
This possibility is a speculative promise that we could learn about Heidi’s dreams. The pieces are not all in place yet. Just as hearing about a dream is not the same as dreaming the dream, the experience of an animal’s dream will always be subjective. Still, behaviors and brain activity that accompany dreams could allow us a little bit into the world of animal dreams.
The brain’s reward system is the same system that drives dreaming—it is integral to motivation. This circuit is also known as the wanting or seeking system. It has a large role in foraging behavior.
I speculated that Heidi might be dreaming, and if she were, she might be dreaming of catching a crab. During attacks on prey, the octopus employs a series of body patterns, different from the patterns seen when a predator attacks an awake octopus. Sufficient ecological work on waking octopuses, octopus-learning experiments, and further sleep studies could reveal whether these body pattern sequences when asleep might be understood as dream-enacting behaviors. This could reveal both octopus nightmares and octopus dreams.
______________________________
Many Things Under a Rock by David Scheel is available via Norton.
