Your environment is making more decisions than you're

In 1986, a team led by Charles Czeisler showed that bright light could reset the human body clock, whatever hours the person slept.1 That clock is the circadian pacemaker, the timer that sets when you feel awake and when you feel sleepy. It wasn't mainly tracking bedtime, or routine, or resolve. It was tracking light.

The finding has held up: researchers took people camping for a week with no electric light. Their inner timing moved earlier, and so did their melatonin, the hormone the body puts out before sleep. Its release moved closer to sunset.2 Nothing about the people had changed. Their conditions had.

Run it backwards and the mirror image appears. Several evenings of reading on a lit screen before bed pushed melatonin later and pushed sleep later. People were less alert next morning than after a printed book.3

Now consider how this feels from the inside. Each evening you choose when to go to bed, and it has the full texture of a choice. But much of the answer was settled hours earlier, by the lamps and the screen in your hand. It came through a pathway you can't feel and never agreed to.

Here's the claim I want to make, and a thoughtful reader can disagree with it. The self that makes choices isn't sealed off from the room it's standing in. Light, noise, temperature, layout, distance, social norms and the way screens are designed aren't the backdrop to what we do. They're inputs to it, and they work mostly below the level where thought happens. This isn't an argument that people lack agency. It's an argument about where agency does the most good. Choosing your conditions is far more powerful than choosing your actions inside conditions you didn't design.

The exposure you can hear

Light is the cleanest case, because the pathway isn't in doubt. Noise is more instructive, because of how badly it gets misfiled.

Noise is usually treated as a matter of taste. Some people mind the road, some don't, and the ones who mind should relax. Research treats it as an exposure instead, much as it treats dirty air. A major review in The Lancet set out effects far from the ear: broken sleep, weaker thinking in children, links to heart disease.4 It also treated annoyance as a health outcome in itself. A later review in a heart journal traced the paths: stress hormones, chemical damage inside cells, harm to blood vessels.5 Night-time noise was singled out as the worst, because the body answers sound without needing you awake.

"I've got used to it" is among the most confident claims in this field, and among the least sound. Not noticing a sound isn't the same as your body not answering it. The two are different processes. Someone can stop hearing the motorway past the bedroom window while their blood vessels carry on answering it, night after night, for a decade.

This is what marks off an exposure from most of what gets called lifestyle. Diet has gaps. Exercise has rest days. Work stops, at least in theory. Your conditions have no off state, and the meter runs whether or not anyone reads it.

What green space can and can't prove

The obvious answer to a bad setting is a good one, and the best studied good one is nature. Here we need more care than most coverage brings.

Stephen Kaplan proposed an idea called attention restoration. Directed attention is the effortful kind, used to hold one thing in mind while shutting others out. Kaplan argued that it wears out, and that natural places let it recover through what he called soft fascination.6 There's enough interest to occupy attention, not enough demand to consume it. Berman and colleagues found that a park of trees beat city streets on a test of directed attention.7 Bratman and colleagues found that ninety minutes in nature reduced reported brooding, compared with a walk beside traffic. Activity also fell in a brain region linked to brooding.8

Scale it up and the picture blurs. A survey of nearly twenty thousand people in England found a threshold. Those spending at least two hours a week in nature more often reported good health and high wellbeing. Below two hours, no such link appeared.9 Headlines ran the number as a dose, which it isn't. The study photographed a population at one moment, so it can't tell us which came first. People who spend two hours a week in green space differ from those who don't. Money, housing, mobility and whether there's anywhere green all predict health too. The major review of the field is candid: several paths could connect nature to health, among them cleaner air, walking and social contact. Separating those from each other, and from money and housing, is really hard.10

The immune findings need more care still. Qing Li reported that trips into forests raised the activity of natural killer cells, a type of immune cell.11 The rise lasted days. The result gets cited as though it settled something. The samples were small. Nobody was blinded, so people knew which group they were in, and that can change what they report. There was no proper control group either, which is hard to arrange, because you can't hide a forest from the person standing in it. This is real work at an early stage. Interest is the right response, not confidence.

None of this dissolves the case; it changes its shape. Natural places do seem to restore something over short spells, and the links turn up in many countries and study designs. But the causal path isn't pinned down, and anyone telling you otherwise is usually selling a retreat.

Habits live in places

If the first argument was about what places do to bodies, this one is about what they do to choices.

The folk model says people decide things and then do them, and that when the doing fails, the deciding was weak. Research on habit describes something else. Wood and Neal argued that a well practised action is set off directly by recurring features of the setting.12 That means the place, the time of day, the act just before, the people present. Call this a context-cued habit: a habit fired by the setting rather than by a fresh choice. What people say they intend predicts what they do fairly well while a habit is weak, and much less well once it's strong.

This is why the same person can be steadily disciplined in one setting and not in another. Nothing changed in their character between Tuesday and Saturday. It's also why moving house breaks habits so well: a move takes apart the cue structure that was running them. Behaviour that survived years of resolve can collapse in a fortnight, because the kitchen is a different shape.

Marteau and colleagues drew the public health lesson. Aim at the automatic side of behaviour, mainly by changing the places where choices are made.13 That may achieve more at population scale than appealing to conscious thought. They were explicit that the evidence for such measures was thin. That caveat hasn't been resolved; it's become a live debate. Such changes often prove weaker in the field than in the lab. When an action keeps failing, the useful question is about the setting rather than the person. What, exactly, is this set-up built to produce?

A pattern inside a larger pattern

There's an older way of describing a human being that makes all of this obvious.

A person isn't a bounded unit sitting inside a neutral box, but a pattern held inside a larger pattern. You trade light, air, sound, food and attention with everything around you, and that trade keeps reshaping you. On this reading, health belongs to a relationship, not to the body alone. A plant's condition belongs to soil, water and light. Move the plant and the same plant does differently. Nothing about it got weaker.

Ecology supplies a useful concept, with a warning. In a trophic cascade, a change in one link of a food web spreads through the whole system, reshaping parts nothing touched directly. Yellowstone after the wolves returned is the famous case, and it's been badly oversold. In the popular form, wolves changed the course of rivers, which claims far more than the evidence supports. The fifteen year review of the plant data found regrowth that was real but patchy, and uneven across sites. It was tangled up with climate, elk numbers and older land use.14 Later work has cast more doubt on its size. The concept of a cascade is sound. The size of that one is contested: serious researchers disagree about it. Use it as an example, not as proof.

Systems have load-bearing links. Change one of those and the effect can be far larger than the effort. Most self-help works downstream instead, where the effect is far too small. Restoring one link the whole set-up leans on is a different move, and that link is rarely the one attracting attention. It's often something dull and structural: what time the light goes off, or who else is in the house.

The mismatch argument, held loosely

The obvious next step is to say that modern conditions are wrong for the animal. That step is where the argument becomes easiest to overstate.

Humans took shape while moving nearly every day, under light that swung hugely between noon and midnight. Heat swung with the season, food came and went, and groups were small, steady and in the room with you. Modern life holds those dials strangely still. The light is much the same at nine in the morning and nine at night, the temperature the same in February and July. Under this reading, much of what modern life produces isn't the effect of one harmful input. It's the effect of sameness, in a system built to be tuned by change.

I find that a strong argument, but it's our reading, not a settled fact. Evolutionary mismatch reasoning is famously hard to test. The ancient setting can't be observed, and the modern one can't be taken away in an experiment. Worse, used loosely it can't be proved wrong. Any modern complaint can be paired with any modern novelty and made to sound like an answer. That pairing has produced much confident nonsense.

What splits the sound form from the unsound one is mechanism. The body clock case is strong because it doesn't rest on a story about ancestors. It rests on a traced path from light to the timing of a hormone, measured in a lab within a week.2 Where a mismatch argument comes with a traced path through the body, treat it as evidence. Where it comes only with a story about the savannah, treat it as a guess that hasn't been asked to do any work.

Where the leverage actually sits

Two objections deserve answers before this argument can stand.

The first is determinism, the idea that surroundings dictate what people do. It fails. Surroundings load the dice, and they load them unevenly. People differ hugely in how much the same setting affects them, and much of that spread is unexplained. The claim is about how influence is shared out, not that it's total.

The second objection is harder, and it's political. Much of what shapes a person's conditions isn't theirs to select. Whether the flat faces a busy road. Whether there's anywhere green within walking distance. These track income, housing markets, job terms and planning choices made decades ago by people who will never meet the tenant. The nature research is tangled up with exactly this, which isn't an accident. Hand someone the finding that a setting governs behaviour, then make fixing it their private duty, and you've traded one moralism for another. "You lack discipline" becomes "you lack a well-designed home." That's not obviously the kinder charge.

What survives both objections is narrower and more useful. Two kinds of choice are open to a person. There are choices made inside a setting: what to eat now, whether to sleep now, whether to open the app. These recur endlessly, they arrive when your capacity is lowest, and they're exactly what a room can bias. Then there are choices about the setting. Where the phone spends the night. What lights the last hour before bed. Whether the route to work passes anything living. These are rare, made once, usually while thinking clearly, and then they run by themselves for months.

The second kind is small, dull, and holds most of the leverage. This isn't because effort is fake or discipline a myth. Effort is a recurring cost, drawn from a budget already heavily spent. Conditions, once arranged, carry on working while you sleep.

Which leaves one thing, and it has nothing to do with lamps or trees. Suppose someone's conditions have held more or less constant for years. The same rooms, the same light, the same commute, the same sound at three in the morning. Then everything they've learned about themselves has been learned at one setting of the dials. The tiredness, the appetite, the focus that fails after lunch, the person who shows up in meetings. All of it measured once, in one place, with nothing held apart to compare it with. Most of what gets called self-knowledge is knowledge of one animal in one habitat, and the habitat has never been varied. Which means the honest answer to what am I actually like may be that nobody has yet run the experiment.