WAIT, before anything else–today's the day! Robot-Proof: When Machines Have All The Answers, Build Better People is out today! (Assuming your "today" is March 17, 2026. If not, hello future...or past?) Go buy a copy before you do anything else today. Buy on Amazon. Or buy on Barnes & Noble or Porchlight, Bookshop, or BAM. Or buy a signed copy from me at Socos Labs.

Now back to your regularly scheduled newsletter:

Your brain is not a fixed asset. It's an organ that responds to the demands you place on it—build reserve or burn it down, one thought at a time. This week we’ll explore 3 papers that together explain why what you think about matters far less than how you think.

Research Roundup

Five years. No drugs. No surgery. Just a full life.

That's the headline finding from one of the most important longitudinal studies in cognitive aging: the Rush Memory and Aging Project, which followed nearly 2,000 older adults in northeastern Illinois for an average of 7.6 years — tracking them until death, and then, for a subset of nearly 1,000, examining their brains.

Does lifetime cognitive enrichment predict who gets dementia and when?

Lifetime cognitive enrichment isn’t a snapshot of what you're doing today; it’s a picture of how cognitively demanding your life has been across decades: your education, the complexity of your work, how socially and intellectually engaged you've kept yourself. Measured this way, the answer was blunt…

…1 unit higher in lifetime enrichment was associated with a “38% lower hazard” of developing Alzheimer's dementia. People in the top 90th percentile of lifetime enrichment developed the disease an average of “5 years later” than people in the bottom 10th. In a disease where every approved pharmacological intervention struggles to demonstrate even modest symptom delay, 5 years is enormous.

But the real finding—the one that should reframe how you think about cognitive aging—came from the autopsied subset. When researchers examined the brains of participants who had died, they found something counterintuitive: lifetime enrichment did not meaningfully reduce the physical pathology of Alzheimer's disease.

The plaques were there. The tangles were there. Enrichment hadn't cleaned the brain. What it had done was preserve function despite the damage. People with high lifetime enrichment maintained better cognitive performance closer to death and showed slower cognitive decline even after statistically adjusting for the burden of neuropathology in their brains.

The technical term for this is cognitive resilience: the brain's ability to absorb damage without proportional functional loss. And it turns out you spend a lifetime building it, or not.

This matters enormously for how we think about the relationship between enrichment and brain health. The naive model says: do cognitively stimulating things, prevent neurodegeneration. The actual model is more interesting and more demanding: do cognitively stimulating things, and your brain develops greater capacity to withstand neurodegeneration when it comes. You are not preventing the storm. You are building a better structure.

One clarification worth making: "lifetime cognitive enrichment" in this study is a composite — it isn't IQ, it isn't formal education alone, and it certainly isn't any single intervention. What seems to matter is the cumulative pattern of cognitive demand across a life: jobs that require genuine problem-solving, relationships with genuine intellectual complexity, leisure activities that require you to generate rather than just consume. The implication is both hopeful and uncomfortable. Hopeful because this reserve can be built at any stage of life. Uncomfortable because not all cognitive activity counts equally…and much of what fills our days and screens probably doesn't count much at all.

Your brain’s environment changes its biology

This is what the enriched environment literature has been demonstrating in animal models for decades, and what a comprehensive review of the neurodevelopmental and neurological repair literature now makes difficult to dismiss in humans: the brain is not a fixed organ running its predetermined genetic program. It is a structure that remodels itself in response to the demands placed on it—growing new synaptic connections, increasing dendritic branching, generating new neurons in the hippocampal dentate gyrus—and it does so across the entire lifespan, not just in childhood.

The review covers "enriched environment (EE)" research across developmental stages—prenatal, pre-weaning, adolescence, and adulthood—and the findings are not uniformly reassuring. Some of the most important results are about limitations and timing, not just benefits.

The first important nuance: not all components of an enriched environment produce the same effects at the same time. Physical activity, for instance, does not produce the neuroplastic benefits of EE before weaning—the animal simply doesn't respond to it yet. Cognitive novelty and social complexity, by contrast, show effects earlier. This means enrichment isn't a single undifferentiated good you can apply at any dose at any point. The developmental window matters. What builds reserve at age 25 is not identical to what builds it at 65, and what you miss in one period can only be partially compensated for in another.

The second finding cuts even deeper: EE is not always superior to its individual components. Physical exercise alone sometimes matches or exceeds full environmental enrichment on specific neurobiological measures. This challenges the assumption that more complexity always means more benefit. What seems to matter is not maximal environmental stimulation but the right kind of demand at the right time, engaging the right systems. Don’t spend every moment in deep processing, but develop a healthy dynamic moving from shallow to deep though a day.

Then there is the neurological repair dimension, which the review examines through the lens of cerebral ischemia. After brain injury, enriched environments accelerate functional recovery, restoring movement, cognition, and behavioral competence faster than standard conditions. But crucially, EE doesn't always reduce infarct size—the damage is the damage. What changes is the brain's capacity to route around it, compensate for it, and recruit alternative networks. This is the same resilience logic that appeared in the first paper’s autopsy findings: enrichment doesn't erase the pathology; it builds the architecture to survive it.

There is one finding in this literature that I find underappreciated, and it sits at the heart of what I want to argue in the coming weeks. In animal EE studies, not every animal in the enriched environment benefits equally. Individuals who actively investigate the novel objects, engage with the social environment, and explore the new structures show greater neuroplastic gains than individuals who are present in the same environment but engage with it passively.

The environment is the exogenous condition. The engagement is endogenous. And the neurobiology responds to the engagement, not the environment.

A mouse can live in the most cognitively enriched cage ever designed and learn almost nothing, if it doesn't engage. A person can sit in a lecture hall at the world's greatest university or scroll through the most intellectually sophisticated content feed ever assembled…and build essentially no reserve. If the cognitive processing is shallow, passive, and non-generative, you gain nothing.

An enriched environment is necessary but not sufficient. What you do within your environment is the variable that the neuroplasticity literature is actually tracking, and it is the variable that is most at risk when a tool arrives that is perfectly designed to do your thinking for you.

SuperAgers exist—so let’s crack open their brains.

So (1) cognitive enrichment across a lifetime reduces the risk of Alzheimer's dementia (AD), delays its onset, and builds resilience against its pathology—not by preventing neurodegeneration, but by building the capacity to absorb it. And (2) genuine cognitive demand, not mere exposure, drives neuroplasticity, with changes visible in synaptic density, dendritic architecture, and hippocampal neurogenesis.

Let’s ask a harder question: can we see the signature of cognitive resilience not in behavior, not in neural architecture, but in the molecular activity of individual cells? The answer appears to be yes.

A new study examined post-mortem hippocampal tissue from 5 groups:

1. young adults with intact memory,
2. older adults with no cognitive impairment,
3. SuperAgers, adults in their 80s and beyond with memory performance matching people decades younger
4. adults with preclinical or intermediate AD pathology, and
5. adults with diagnosed Alzheimer's disease.

Using multiomic single-cell sequencing, it analyzed the profiles of 355,997 individual nuclei, characterizing neural stem cells, neuroblasts, and immature granule neurons across all 5 groups.

What they found in the SuperAgers was not simply more neurons, bigger hippocampi, or higher rates of neurogenesis. It was a qualitatively distinct profile of neurogenic activity: a specific pattern of gene expression and chromatin accessibility that doesn't appear in normal aging and doesn't appear in any of the AD groups. The researchers describe it as a potential resilience signature—not just the absence of decline but the presence of something actively protective operating at the level of gene regulation.

Chromatin accessibility is essentially a measure of which parts of the genome are available to be read (i.e. which genes are open for transcription and which are effectively silenced). What the researchers found is that dysregulated neurogenesis in Alzheimer's disease is largely driven not by structural brain damage but by changes in chromatin accessibility, epigenetic changes that alter which genes hippocampal stem cells can express. Critically, these chromatin changes appear in preclinical AD (individuals with early pathological signs but no clinical symptoms yet) before the cognitive decline that would bring someone to a memory clinic.

The failure of neurogenesis in Alzheimer's begins, at the molecular level, before anyone knows they're sick.

This reframes the entire timeline of intervention. If the genomic architecture of hippocampal resilience is being set, or being compromised, years before any clinical presentation, then the question of when cognitive enrichment matters shifts dramatically. The window is not "once symptoms appear" or even "in middle age". The chromatin accessibility data suggests that the molecular conditions for resilience or vulnerability are being written continuously, across an entire lifetime, through mechanisms that are responsive to experience.

Which brings us to what may be the most important implication of this paper, and the one I'll be returning to in Thursday's paid newsletter.

The SuperAger resilience signature is not genetic destiny. These individuals didn't simply inherit protected hippocampi. The epigenetic nature of the finding—changes in chromatin accessibility, transcription factor activity, and the regulation of neurogenic cells—points toward a system that is written by experience. The molecular machinery of cognitive resilience is, at least in part, shaped by what the brain is asked to do.

We have now followed the thread from behavior to mechanism to molecule. At every level of resolution, the same logic holds: cognitive demand builds the architecture that withstands decline. The question that follows—the one that scares me—is what happens to that architecture when the demand is quietly, systematically removed.

Media Mentions

SciFi, Fantasy, & Me

Arkady Martine's A Memory Called Empire is a diplomatic thriller set in a vast, consuming empire — and also, unexpectedly, the most precise fictional treatment I've encountered of what it means to think with a tool versus think through one. Its central technology is a recorded predecessor personality implanted in their successor: when it works, the protagonist is sharper, faster, and more politically fluent than she could ever be alone; when it fails, she realizes she can no longer distinguish her own reasoning from the reasoning she inherited. Hugo Award winner; read it slowly.

Stage & Screen

• March 17, Online: The book launch! Robot-Proof: When Machines Have All The Answers, Build Better People is will finally be inflicted on the world.
• March 27, Geneva: Its a secrect, but I'll tell you soon.
• March 30, Amsterdam: What else: AI and human I--together is better!
• April 2, Paris: A book reading on the Seine!
• April 14, Seattle: Ill be keynoting at the AACSB Business School Conference.
• April 16, NYC: A private event in Brooklyn. The topic is AI, but I'll make it about us.
• May 12, Online: I'll be reading from Robot-Proof for the The Library Speakers Consortium.
• May 12, SF: We'll talk about collective intelligence, the neuroscience of trust, and how dumb I have to be to be launching my 13th company.
• May 14, Miami: TEDxMiami
• June 9-10, London: London Tech Week!
• June 11, Luxembourg: How Europe (and even some of it smallest states) compete and grow in a trade environment dominated by zero-sum leaders
• June 12, Denver: GlobalMindEd
• June 18, Stockholm: The Smartest Thing on the Planet: Hybrid Intelligence
• October, Toronto: The Future of Work...in the Future

Vivienne L'Ecuyer Ming