Your brain isn't broken — it's running ancient software

You have probably run the diagnostic on yourself already. Can't focus for more than a few minutes. Tired in a way sleep doesn't fix. Low for no reason you can point at. Restless in a room full of people you like.

And the conclusion is nearly always the same: something is wrong with me.

Here is a different reading. Your brain is running a build that is roughly 300,000 years old, and it has never been updated. Not for offices. Not for cars. Not for a screen that fits in your hand. The hardware is fine. It is running in an environment it was never written for, and almost everything you call a symptom is the gap between the two.

Both things can't be true

If the machine didn't change, what did?

Either millions of people independently developed the same defect, at the same moment in history, in the same handful of countries — or the room they're standing in changed and the machine didn't.

One of those is a coincidence the size of a civilisation. The other is just history.

The clock

Our species is about 300,000 years old. Farming is 12,000. Industry, 250. A majority of American adults have only owned a smartphone since around 2013 — about thirteen years at population level.

Put that on one line and the shape is hard to miss. Roughly 96% of the human story was spent in small mobile groups, outdoors, on foot, in daylight, eating things you could recognise. Everything you think of as ordinary life fits inside the last 4%.

One honest wrinkle: the fine-tuning ran later than that headline number suggests — brain shape kept changing until perhaps 35,000 years ago, even though brain size was there much earlier. But nothing on that timeline is recent enough to matter here. Four or five hundred generations since farming is nowhere near enough time to redesign a brain.

So the useful question isn't what's wrong with you. It's narrower: what did that build expect to receive every day that it now doesn't?

Four things, each with a real number attached.

One: movement

The average Western adult now spends more than half the waking day sitting — accelerometer data puts it near 57%.

You know what usually comes next: sitting is the new smoking. That line is wrong, and it's worth knowing why. The largest analysis of it followed over a million people. Sitting does carry real risk, and the risk climbs with each hour past about seven. But 60 to 75 minutes of moderate activity a day essentially erases the excess mortality. Smoking works nothing like that — you cannot undo a pack a day with an hour's walk.

The honest version is less quotable and far more useful: sitting is a dose you can offset, and most people simply aren't offsetting it.

How much offsetting? Less than you've been sold. The 10,000-steps figure came from a 1960s Japanese pedometer campaign, not a laboratory. When researchers measured it properly — fifteen cohorts, some 47,000 people — the mortality curve bent early. Against the least active group, about 5,800 steps a day cut risk by roughly 40%; 7,800 by about 45%. After that the curve flattens, and for adults over sixty it flattens sooner.

For contrast, the Hadza, who still live by hunting and gathering, average around 15,000 steps a day and show almost none of our cardiovascular markers — though they also eat differently and face different disease pressures, so movement isn't the only variable there.

And most of what your body burns isn't exercise at all. Ordinary standing, walking and fidgeting can swing daily energy expenditure by as much as 2,000 calories between two people of the same size.

Two: light

Americans spend about 87% of their time indoors and another 5.5% in vehicles. Call it nine tenths of waking life under a roof.

That matters because indoor and outdoor light are not the same thing wearing different clothes. A well-lit room is somewhere between 50 and 500 lux. An overcast day outside is 1,000 to 10,000. Direct sun runs past 100,000. Even a grey, miserable afternoon outdoors is brighter than the best-lit room in your house, usually by a factor of ten or more.

Here's what happens when that light arrives. In a cluster-randomised trial in China, researchers added 40 minutes of outdoor time to the school day and followed nearly 1,900 children for three years. Myopia developed in 30% of the children who went outside, against about 40% of those who didn't. Same classrooms, same books. The difference was daylight.

Morning light also nudges your internal clock earlier — though the shift per morning is minutes rather than hours, and it builds with consistency rather than arriving overnight.

And one correction while we're here. You've been told blue light before bed wrecks your sleep. The mechanism is real in a laboratory; the product built on it isn't. A meta-analysis of blue-blocking glasses found small, inconsistent effects with confidence intervals crossing zero. Dim everything and go to bed on time. Don't buy thirty-dollar glasses and call it solved.

Three: food

About 58% of American calories now come from ultra-processed products. Is that actually doing anything, or is "ultra-processed" just a word people disapprove of?

In 2019 researchers at the National Institutes of Health answered it properly. Twenty adults lived in a research ward for a month: two weeks eating ultra-processed food, two weeks eating whole food. The two diets were matched on paper for calories, sugar, fat, sodium and fibre. People could eat as much or as little as they liked.

On the ultra-processed diet they ate 508 calories a day more and gained about a kilo in two weeks. On whole food, the same people lost about a kilo. Researchers think part of the reason is that these foods are eaten faster and pack more calories per bite, though the exact mechanism is still being worked out.

What that experiment does not license is the story usually bolted onto it. There is no such thing as the ancestral diet. Across hunter-gatherer societies, carbohydrate ran anywhere from a fifth to over half of daily calories and fat from 12% to nearly 60%, depending entirely on latitude and ecology. Nobody was doing keto on principle.

What's consistent across all of them isn't a macronutrient ratio. It's the absence of anything engineered to be eaten quickly.

One thing does stand out: estimates put Hadza fibre intake at 80 to 150 grams a day. Those are reconstructed figures rather than direct measurements, so hold them loosely. Even at the low end it's five to ten times the American average of about 15 grams.

Four: company

In 2003, young American adults spent about 61 minutes a day with friends. By 2019 — before anything closed — it was 39. A third of the time gone in sixteen years, with the pandemic year a separate story stacked on top.

Meanwhile the places that generated contact without anyone planning it have thinned out. When Microsoft studied 61,000 of its own employees through the shift to remote work, the finding wasn't that people stopped talking. It was that the network went rigid: more siloed, more asynchronous, with fewer of the accidental cross-group ties that make an organisation feel like a place rather than a directory.

The health side is where the numbers get serious and also where they get mangled. You've heard that loneliness is as bad as smoking fifteen cigarettes a day. That comes from a review of 148 studies covering 300,000 people, which found people with stronger social ties had a 50% higher likelihood of survival across the study periods.

But the cigarettes line is a comparison of risk size, not biology. Researchers weren't claiming loneliness gives you the same diseases as smoking — they meant the two take roughly the same amount off your odds. The lead author has said publicly that the statistic became a game of telephone.

The quieter finding is more usable anyway: about 60% of your daily social contact is with weak ties — the barista, the neighbour, the person at the next desk — and more of that contact predicts a greater sense of belonging.

Look at the symptom list again

Scattered attention. Flat mood. Restlessness. Loneliness in a crowded city.

Those aren't malfunctions. That's a well-built machine reporting, accurately, that four of its operating conditions are missing.

You are not broken. You are running ancient software on a system it was never written for, and it is doing its best to tell you so.

Which is oddly good news, because environments are editable and personalities mostly are not. You cannot decide to want different things. You can change the room.

The smallest intervention that touches the most inputs

Twenty minutes outside, in the morning, on foot, ideally with someone.

That's one habit, and it pays into all four. Daylight, at an intensity no room can match. Movement, of the ordinary kind the step data says matters most. Company if you bring a person — and a weak tie or two if you don't. It does nothing directly about food, so pair it with the only food rule the evidence really supports: eat things that take time to eat.

It's worth being precise about what that claim is worth. Exercise has a large measured effect on depressive symptoms compared with no treatment, but head-to-head against antidepressants the difference isn't significant — so this is not a replacement for treatment, and nothing here says otherwise. Morning light reliably shifts the clock, but no trial has tested a twenty-minute walk specifically, measuring mood and sleep and metabolism together.

Researchers haven't tested this exact recipe as one intervention. What they have shown, again and again, is that stacking these inputs together beats doing them in isolation. This is the simplest way we could find to stack four of them into one habit.

You were never the problem. The room was. And the room is the part you're allowed to change. Pick one input, put it back this week, and see what your hardware does with it.

The full argument, with every source on screen, is in the film — watch it on YouTube.