Your Thinking Is Atrophying. Here's What I'm Doing About It.

I used to know phone numbers by heart. Not a few. All of them.

I had a Nokia 3310, I don't even remember what year that was..and yes, it had a contact list. But I knew the numbers anyway. My friends, my family, the people I called regularly. Maybe 80 to 100 numbers, sitting in my head, available instantly.

I couldn't tell you a single one today. Not my parents'. Not my closest friend's. And for a long time I didn't think about it. Why would I? The phone knows. That's the point. That's progress.

I know I’m not alone.

When was last time you navigated somewhere without Google Maps, actually read the environment, asked someone, took the wrong turn, and were ok with it? When you have a question you don’t immediately know the answer to, how long do you last before reaching for your phone instead of thinking it through yourself?

And then came the realization I didn't want to admit.

I am getting dumber.

And so are you.

This isn't personal, and it isn't a discipline problem. It isn't about screens or laziness or willpower either.

It's measurable. It's documented. And the data is worse than I expected.

So I went looking. I wanted to know if what I was feeling was real or if I was just being dramatic about a phone. It was real. IQ scores have been measured since the early 1900s. And for almost a century, they followed a remarkably stable pattern: every generation scored higher than the one before. Researchers called it the Flynn Effect. It held through wars and recessions. The line kept climbing.

Until now.

For the first time since they started measuring, scores are going down. Not plateauing. Actually reversing. A generation scoring lower than the one before it. Something that had never happened.

And the researchers trying to explain it kept landing on the same things: less sustained attention (the ability to stay with one thing without switching), more cognitive outsourcing (letting your phone or AI do the thinking instead of you), and fewer moments where people are forced to sit with something hard without immediately reaching for an escape hatch (that reflex to google something the second you don't know the answer).

I found that interesting, because it didn't start with AI. It didn't even start with the smartphone. It started way earlier, and it happened so gradually that nobody thought to question it because every single step felt like progress.

Calculators meant we stopped doing arithmetic in our heads. Search meant we stopped retaining information. GPS meant we stopped building internal maps of the world. Smartphones put all of it in one pocket. And now.. AI drafts our thoughts, finishes our sentences, and hands us answers before we've had the chance to even attempt the question ourselves.

Each step feels like an upgrade. Each step taking something with it.

And when I really sat with that (I mean really sat with it), I realized something else.

This has happened before.

Not with thinking. With something else. Something we can only see clearly now because we're standing on the other side of it.

Here's the thing about patterns: they're almost impossible to see from the inside. Let me try something with you..

Let’s hop into the Time Machine for a second.

Imagine it's 1920.

You're a farmer in rural France. You wake up before sunrise, haul water, work the land with your hands for ten hours, walk several miles before noon without thinking about it. Your body is not something you "work on." It's just what carries you through the day.

Now I appear and tell you about the future. "In about fifty years," I say, "most people will sit in chairs all day. They'll move their fingers for a living. They'll drive everywhere, even short distances. Their bodies will barely be used."

You'd probably nod along.

But then I say: "Because of all that, people will start paying money to go to special buildings where they lift heavy objects… and then put them back down. They won't build anything with them. They'll just pick them up, put them down, and go home."

You'd stop me there. "Pick them up and put them back down? The same objects? Every day? And they pay for this?"

Yes. Monthly. Some of them quite a lot.

"But… why not just carry something useful? Why not chop wood, or move stone, or—"

Because there's no wood that needs chopping. The demand disappeared. So they had to rebuild it artificially; on purpose, in a dedicated space, with structured progression, just to keep the body from completely forgetting what it was built to do.

You stare at me: "Ben, I like you and all..but that's the stupidest f*cking thing I've ever heard."

-

And yet, here we are, 50 years later. Gyms are a $100 billion industry. Nobody thinks it's strange anymore.

The pattern is always the same. We remove a friction. A human capacity that depended on that friction atrophies. We eventually build a synthetic environment to train it back (deliberately, structurally, with progressive resistance) because the world no longer provides it by default.

Every single time, the same arc. Every single time, an entire industry was born on the other side.

The supplement industry. The therapy industry. The meditation app industry. The gym industry. All of them built on the same foundation: Something that used to be free and automatic became scarce and precious. And the people who recognized that early (who saw the atrophy coming before it was obvious) built the environments that everyone eventually needed.

We are at that exact moment right now with thinking.

Not in five years. Not when AI gets more powerful. Now. Today.

The friction is already gone. The atrophy is already measurable. The IQ data isn't a warning about the future, it's a report on what's already happening. Which means the industry that's coming is inevitable.

Deliberate cognitive training environments. Structured resistance for specific mental muscles. Progressive overload for the mind: the same logic that built the gym, applied to thinking.

That's not a prediction I'm making nervously. It's the only logical conclusion of a pattern that has proven itself every single time without exception.

The same gap is opening right now.

And I'm not waiting to see who fills it.

I've been building my version for months. Testing it on myself. Refining it. Running it daily. I looked at the pattern, followed it forward, and decided I'd rather be on the right side of this early than scramble to catch up later when everyone else finally sees what's happening:

The atrophy isn't coming. It's here.

The question is no longer whether this is happening. The question is what you're going to do about it.

So here's the stack. The exact practices I'm running right now to make sure my mind is training while everyone else is outsourcing. It will evolve. I'll keep sharing as it does. But this is where I am today.

If you want to stay sharp, this is the direction.

My Cognitive Training Stack (Steal it. Get ahead).

1. The Double Entry Protocol Trains: Original thought, problem solving, sustained attention, belief formation

Writing by hand, from scratch, with no assistance, is one of the most complete cognitive workouts that exists. It forces you to hold an idea in your head, produce original thought before consuming anyone else's, and sit with uncertainty long enough to reason through it (all without autocomplete, without search, without AI finishing the sentence for you).

This protocol is a structured 4-part daily writing practice built entirely around that resistance. Twenty minutes. Every day. No tools.

It's the compound lift of this entire stack. The single practice that trains the most cognitive muscles simultaneously.

It’s also placed in first position, because you can't become a different person with a shrinking mind. But you also can't become a different person by just thinking about it. That's the trap most self-aware people live in: Understanding exactly who they want to become, and still waking up as the same person every day.

If you want the full breakdown of the protocol (and why it works biologically), it’s here.

2. Reverse Recall Trains: Episodic memory, future simulation

Memory is the first thing to go. And most people don't notice, because the phone noticed for them.

It's been storing your days, your appointments, your routes, your contacts, your conversations effortlessly. And every time it did, your brain got the signal: don't bother.

And while you probably already noted the memory loss, what you probably didn’t know is that you don't just lose the past. The same neural machinery that reconstructs what happened is the only machinery that can simulate what's coming. Memory and imagination run on identical hardware. So when you damage one, you degrade the other.

So the practice is simple: every evening, reconstruct your day in reverse order. From memory. No notes. Start with the last thing you did. Work backwards to the moment you woke up.

Start with the last two hours. Add an hour every week. Progressive overload: same logic as the gym.

This is also the practice that makes the Double Entry Protocol work at full capacity. You'll understand why the moment you're doing both.

3. Deliberate Open Loop Trains: Tolerance for ambiguity, original thought

Google didn't just make you lazy. It made you intolerant of your own thinking process.

Every time you reached for your phone to resolve something uncomfortable, you skipped the part where your brain was supposed to do the work. Do that enough times and you stop trusting your own mind to figure things out, because you never give it the chance.

Every morning, you write down one thing that's genuinely sitting with you. It has to be something unresolved. Something like:

Why did that conversation leave me feeling off? What is it about this decision that I keep avoiding?

You don't answer it. You don't google it. You don't ask anyone.

You carry it. Through your morning, your work, your day. Let it run in the background without forcing it.

Tolerance for ambiguity is one of the strongest predictors of creative and strategic thinking on record, because sitting with it is what forces your brain to work things out on its own.

This one looks the least productive on the list. It’s not. Quite the opposite.

Fun fact: learning to sit with uncertainty has also been one of the most effective things I've ever done for my cortisol. An embarrassing number of people in my life have told me I'm "really chill." This is why.

4. First Principles Strip Trains: Assumption removal, bedrock thinking

Every tool that makes you think faster is also making you better at operating inside boxes you didn't build and never questioned.

Search gives you better answers to the question you asked. It never asks if the question is wrong. AI does the same thing: faster, more convincingly, more dangerously I would argue. The whole architecture of modern thinking is optimized for speed inside existing assumptions.

So the capacity that's dying quietly isn't just memory or attention. It's the ability to step outside the frame and ask: why does everyone believe this is true?

Once a week. One belief. Strip it to what you can actually verify.

Not what you were told. Not what everyone assumes. Not what worked before. What is actually, provably true if you start from zero. (if you want the longer version of how to do this, it’s in Chapter 8 of Unf*ck Your Thinking. If you’re read it already, you have a massive advantage).

Write the belief at the top of the page. Ask "how do I know this is true?" Keep asking until you hit something verifiable, or realize you've been standing on an assumption you inherited from someone else and never once examined.

This one is weekly, not daily. (The more you practice, the more you will start doing this naturally).

5. Analog Skill Acquisition Trains: Pattern recognition, perceptual learning, critical thinking

There is a category of learning that does something no other practice on this list can replicate.

When you go deep enough into a complex analog skill (something that requires your hand, your eye, your ear, and your brain to work in coordination, with no shortcut and no way to fake progress) you eventually find an invisible architecture. A construction that was always there but completely invisible from the surface.

And here's what makes this specifically valuable for thinking: once you learn to see underneath one thing, the habit transfers. You start finding the architecture underneath everything. Problems. Assumptions. Systems. Ideas that everyone else is experiencing only at surface level.

That's not a metaphor. That's how perceptual learning actually works.

I chose drawing. More specifically, I've been learning 3 point perspective. What it does is force you to stop seeing a building and start seeing vanishing points, horizon lines, and spatial relationships that were always there but invisible until you had to construct them by hand. Not for the skill itself. For what going underneath it does to the way I see everything else.

Pick something genuinely hard, analog, and unoutsourceable. Go deep enough to find what's underneath the magic.

-

The people who will get ahead in the next decade aren't going to be the ones with access to the best AI.

Everyone has access to the best AI.

They're going to be the ones who kept their minds sharp while everyone else was outsourcing theirs. Who trained the capacities that can't be automated. Who saw the atrophy coming and did something about it before it became obvious.

That's the gap. And it's opening right now.

If you just implement the first one on the list, it is a biological inevitability that you will be ahead. If you implement all of them, you'll be early to the most obvious industry nobody's naming yet, and you'll have already been training for it.

Stay smart my friend,

Benoit