During most of the night, there is no dreaming. But during REM (rapid eye movement) sleep, something special happens. The heart rate and breathing speed up, small muscles twitch, and the brain waves become smaller and faster. This is the stage of sleep in which dreaming occurs.41 REM sleep is triggered by a particular set of neurons in a brainstem structure called the pons. The increased activity in these neurons has two consequences. The first is that the major muscle groups become paralyzed. Elaborate neural circuitry keeps the body frozen during dreaming, and its elaborateness supports the biological importance of dream sleep; presumably, this circuitry would be unlikely to evolve without an important function behind it. The muscular shutdown allows the brain to simulate world experience without actually moving the body around.
The second consequence is the really important one: waves of spikes travel from the brainstem to the occipital cortex.42 When the spikes arrive there, the activity is experienced as visual. We see. This activity is why dreams are pictorial and filmic, instead of conceptual or abstract.
During dream sleep, waves of activity begin in the brainstem and end in the occipital cortex. We suggest this infusion of activity is necessitated by the rotation of the planet into darkness: the visual system needs special strategies to keep its territory intact.
This combination crafts the experience of dreaming: the invasion of the electrical waves into the occipital cortex makes the visual system active, while the muscular paralysis keeps the dreamer from acting on the experiences.
We theorize that the circuitry behind visual dreams is not accidental. Instead, to prevent takeover, the visual system is forced to fight for its territory by generating bursts of activity when the planet rotates into darkness.43 In the face of constant competition for sensory real estate, an occipital self-defense evolved. After all, vision carries mission-critical information, but it is stolen away for half of our hours. Dreams, therefore, may be the strange love child of neural plasticity and the rotation of the planet.
A key point to appreciate is that these nighttime volleys of activity are anatomically precise. They begin in the brainstem and are directed to only one place: the occipital cortex. If the circuitry grew its branches broadly and promiscuously, we’d expect it to connect with many areas throughout the brain. But it doesn’t. It aims with anatomical exactitude at one area alone: a tiny structure called the lateral geniculate nucleus, which broadcasts specifically to the occipital cortex. Through the neuroanatomist’s lens, this high specificity of the circuit suggests an important role.
From this perspective, it should be no surprise that even a person born blind retains the same brainstem-to-occipital-lobe circuitry as everyone else. What about the dreams of blind people? Would they be expected to have no dreaming at all because their brains don’t care about darkness? The answer is instructive. People who have been blind from birth (or were blinded at a very young age) experience no visual imagery in their dreams, but they do have other sensory experiences, such as feeling their way around a rearranged living room or hearing strange animals barking.44 This matches perfectly with the lessons we learned a moment ago: that the occipital cortex of a blind person becomes annexed by the other senses. Thus, in the congenitally blind, nighttime occipital activation still occurs, but it is now experienced as something nonvisual. In other words, under normal circumstances, your genetics expect that the unfair disadvantage of darkness is best combated by sending waves of activity at night to the occipital lobe; this holds true in the brain of the blind, even though the original purpose is lost. Note also that people who become blind after the age of seven have more visual content in their dreams than those who become blind earlier—consistent with the fact that the occipital lobe in the late-blind is less fully conquered by other senses, and so the activity is experienced more visually.45
As an interesting side note, two other brain areas, the hippocampus and the prefrontal cortex, are less active during dream sleep than during the waking state, and this presumably accounts for our difficulty remembering our dreams. Why does your brain shut down these areas? One possibility is that there is no need to write memory if the central purpose of dream sleep is to keep the visual cortex actively fighting off its neighbors.
We can learn a great deal from a cross-species perspective. Some mammals are born immature—meaning they’re unable to walk, get food, regulate their own temperature, or defend themselves. Examples are humans, ferrets, and platypuses. Other mammals are born mature—such as the guinea pig, sheep, and giraffe—all of whom come out of the womb with teeth, fur, open eyes, and an ability to regulate their temperature, walk within an hour of being born, and eat solid food. Here’s the important clue: the animals born immature have much more REM sleep—up to about eight times as much—and this difference is especially clear in the first month of life.46 In our interpretation, when a highly plastic brain drops into the world, it needs to constantly fight to keep things balanced. When a brain arrives mostly solidified, there is less need for it to engage in the nighttime fighting.
Moreover, look at the falloff in REM sleep with age. All mammalian species spend some fraction of their sleep time in REM, and that fraction steadily decreases as they get older.47 In humans, infants spend half of their sleeping time in REM, adults spend only 10–20 percent of sleep in REM, and the elderly spend even less. This cross-species trend is consistent with the fact that infants’ brains are so much more plastic (as we will see in chapter 9), and thus the competition for territory is even more critical. As an animal gets older, cortical takeovers become less possible. The falloff in plasticity parallels the falloff of time spent in REM sleep.
This hypothesis leads to a prediction for the distant future, when we discover life on other planets. Some planets (especially those orbiting red dwarf stars) become locked into place, such that they always have the same surface facing their star: they thus have permanent day on one side of the planet, and permanent night on the other.48 If life-forms on that planet were to have livewired brains even vaguely similar to ours, the prediction would be that those on the daylight side of the planet might have vision like us but would not have dreams. The same prediction would apply for very fast-spinning planets: if the nighttime is shorter than the time of a cortical takeover, then dreaming would also be unnecessary. Thousands of years hence, we might finally know whether we dreamers are in the universal minority.
AS OUTSIDE, SO INSIDE
Most visitors to Admiral Nelson’s statue in London’s Trafalgar Square have probably not considered the distortion of the somatosensory cortex in the left hemisphere of that elevated head. But they should. It exposes one of the most remarkable feats of the brain: the ability to optimally encode the body it is dealing with.
We’ve seen so far that changes to sensory inputs (as with amputation or blindness or deafness) lead to massive cortical reorganization. The brain’s maps are not genetically pre-scripted but instead molded by the input. They are experience-dependent. They are an emergent property of local border competitions rather than the result of a pre-specified global plan. Because neurons that fire together wire together, co-activation establishes neighboring representations in the brain. No matter the shape of your body, it will naturally end up mapped on the brain’s surface.
Evolutionarily, such activity-dependent mechanisms allow natural selection to quickly test out innumerable varieties of body types—from claws to fins, wings to prehensile tails. Nature does not need to genetically rewrite the brain each time it wants to try out a new body plan; it simply lets the brain adjust itself. And this underscores a point that reverberates throughout this book: the brain is very different from a digital computer. We’ll want to abandon our notions of traditional engineering and keep our eyes wide open as we move deeper into the neural terrain.
The shape shifting around the body plan illustrates what happens in all sensory systems. We saw that when people are born blind, their “visual” cortex becomes tuned to hearing, touch, and other senses. And the perceptual consequence of the cortical takeover is increased sensitivity: the more real estate the brain devotes to a task, the higher resolution