Neuroscientist David Eagleman explains how the brain’s plasticity allows us to reshape ourselves, why dreams exist to protect our vision, and how AI can either enhance or undermine our cognitive growth depending on how we use it.
How the Brain Constructs Reality
The brain is a three-pound organ locked inside the skull that builds an internal model of the world, and much of what we experience is a construction rather than objective reality.
Eagleman became fascinated with this at age eight after a fall that seemed to last far longer than the physics should have allowed, sparking a lifelong interest in perception.
You Are a Team of Rivals, Not One Self
The brain contains 86 billion neurons organized into competing neural networks with different drives, which is why people often feel internally conflicted, like wanting a cookie while also wanting to stay healthy.
This internal “parliament” means we are not a single unified self but a collection of rival voices, and understanding this helps explain why we sometimes act against our own interests.
The Ulysses contract is a strategy where you constrain your future behavior now to prevent yourself from acting badly later, such as removing alcohol from the house when trying to quit drinking.
Brain Plasticity: The Brain’s Ability to Reshape Itself
Plasticity refers to the brain’s ability to change its structure and hold new shapes, like molding plastic. Every memory, skill, and habit physically alters neural connections.
Humans have far more plasticity than other animals, which is why a child can adapt to any culture or era, but this also means the brain is born “half-baked” and requires proper input during critical periods like early language learning.
Fluid intelligence (the ability to learn anything) dominates early in life, while crystallized intelligence (accumulated knowledge and skills) grows as we age. The brain peaks in raw connectivity at age two, then prunes unused connections to specialize.
Plasticity does not simply decline with age; rather, the brain changes less because it has already learned how to operate in its environment. Change still happens when things are disrupted, such as during the pandemic.
How to Actually Change Yourself
The key to personal transformation is seeking challenge and novelty, staying in the zone where tasks are frustrating but achievable.
Once you become good at something, you should drop it and take on something you are not good at, because this forces the brain to build new pathways.
The anterior mid-cingulate cortex appears to function like a “willpower muscle,” growing larger in people who consistently do hard things they would rather avoid.
Physical changes in the brain are visible in experts: pianists develop larger motor cortex regions for finger control, jugglers show changes in visual processing areas, and medical students show cortical redistribution during exam periods.
Cognitive reserve explains why some people with physical brain degeneration (like Alzheimer’s pathology) show no symptoms: they built enough alternative neural pathways through lifelong social engagement, games, chores, and challenges.
Retirement without intellectual or social engagement accelerates cognitive decline because the brain stops building new roadways.
The Purpose of Dreaming
After millennia of debate, Eagleman’s research suggests the purpose of dreaming is to defend the visual cortex from takeover by other senses.
Because the planet rotates into darkness for half the day, the visual cortex is at a disadvantage. Every 90 minutes during sleep, the brain blasts random activity into the visual system to keep it active.
Harvard experiments showed that blindfolding sighted people for just 60 minutes caused the visual cortex to begin responding to sound and touch, confirming how quickly takeover can occur.
Dreaming correlates perfectly with brain plasticity across species: humans, the most plastic, dream the most, while animals with less plasticity dream less. Infants spend 50% of sleep in REM, decreasing with age.
Even blind animals like the blind mole rat still dream, because the circuitry is so ancient that evolution has not caught up to their loss of vision.
Social Media, AI, and the Developing Brain
Eagleman is a cyber optimist for young people, arguing that the internet vastly expands their intellectual diet compared to previous generations who were limited to local knowledge.
Curiosity-driven learning (asking Alexa or AI questions) is more effective than the “just in case” knowledge taught in schools, because the brain retains information better when motivated by genuine interest.
The concern about social media’s harm to children is difficult to evaluate scientifically because there is no proper control group, and many confounding variables exist.
Vicious friction (busywork, spreadsheets, taxes) should be outsourced to AI, while virtuous friction (thinking through hard problems, creative strategy) should be preserved because that is where real learning happens.
AI can serve as a tutor that helps people become better at things they are curious about, similar to how Alexander the Great learned from Aristotle. The key is engaging with AI’s responses rather than copy-pasting them.
People can use AI to challenge their own beliefs by asking it to be brutally honest and identify blind spots, which is easier to accept from an AI than from another human.
AI is massively creative in generating remixes of existing concepts but cannot yet do selection, meaning it does not know which output will resonate most with humans. Humans must still curate and choose.
AI predicts what will succeed based on past data (the middle of the distribution), but humans are novelty seekers who gravitate toward things at the edge, which is why truly new hits cannot be predicted by AI.
The music industry illustrates this tension: record labels use formulas to create hits, but the formulas eventually expire because audiences always seek the new.
AI and the Future of Being Human
AI and the brain share a superficial resemblance (both use connected units), but they are fundamentally different. AI requires millions of examples to learn, while humans can learn from a single trial.
AI lacks internal experience and only observes human behavior from the outside, which limits its understanding of what it means to be human.
Eagleman predicts a renaissance in live, in-person experiences (theater, concerts, human connection) because people increasingly value seeing real humans, as demonstrated by the continued success of live events despite digital alternatives.
AI relationships may serve as a “sandbox” for practicing social skills, but millions of years of evolution drive humans toward physical connection with other people.
Individual differences in brain wiring mean that some people may be more susceptible to AI addiction, just as different people have different addictive tendencies.
The Spectrum of Inner Experience
People vary enormously in their inner experiences. Aphantasia (no mental imagery) and hyperphantasia (extremely vivid mental imagery) exist on a spectrum, with the population spread evenly between them.
Ed Catmull, founder of Pixar, is aphantasic, as are many of his best artists, possibly because they must pay closer attention to external subjects rather than relying on internal images.
Synesthesia, found in at least 3% of the population, involves blending of senses (letters triggering colors, music producing visual experiences) and represents an alternative perceptual reality rather than a disorder.
These differences rarely translate to differences in capability because tasks can be accomplished through multiple cognitive strategies.
Protecting Against Dementia and Cognitive Decline
To reduce dementia risk, keep the brain active until the day you die by constantly taking on new challenges and dropping activities once they become easy.
Building new neural pathways creates cognitive reserve, so that even as brain tissue degenerates, alternative routes remain functional.
Good sleep, diet, and exercise remain foundational for brain health. Exercise in animal studies increases the production of new brain cells, though this is not yet confirmed in humans.
Social engagement is critically important because other people are the hardest thing for the brain to model, keeping neural circuits active and complex.
Politics, Polarization, and the Future of Connection
Eagleman’s wish for the next decade is the complexification of relationships, meaning learning to see outgroup members as full humans with their own internal models, even when disagreeing with them.
Dehumanization of outgroups (homeless people, political opponents, drug addicts) dials down social circuitry in the brain, making it harder to see them as people.
Echo chambers are not new, but the internet at least exposes people to the existence of diverse viewpoints, which is preferable to state-controlled single narratives.
There is a market opportunity for a social media platform designed around building genuine human connections rather than maximizing engagement through dopamine-driven algorithms, though the economic incentives of ad-driven models make this difficult.
Eagleman is ultimately optimistic that humans will seek meaningful connections over pure dopamine hits, just as people do not eat cookies all the time despite their appeal.
