The Architecture of Self-Sabotage
This is a three-part series. You’re reading the long-form analysis. Also available: the interactive data journalism version and the full source audit with bias analysis.
Look — I’ve put off doing this one a little bit. Not because the data is hard to find. It’s everywhere. It’s because the picture it paints is, frankly, infuriating.
The systems that shape your daily life — how your streets are built, what food is cheapest, which content appears on your screen, where you gather (or don’t) — are quietly optimized against you. Against your health, your attention, your relationships, your ability to think clearly. And I don’t mean that in some conspiracy-brained way. This is incentive alignment. Corporations profit when you consume more, move less, and engage passively. Governments built infrastructure around cars, not bodies. The food system subsidizes corn syrup, not broccoli.
The result? The path of least resistance leads reliably toward obesity, isolation, cognitive fragmentation, and chronic disease. Doing what’s actually good for you requires constant, deliberate swimming upstream.
The evidence for this is not anecdotal. It is structural, measurable, and staggeringly well-documented. And the good news — buried under decades of misguided emphasis on individual willpower — is that the handful of communities and countries that redesigned their defaults are seeing dramatic results.
The streets were built for cars, not for living
The most basic form of human exercise — walking from one place to another — has been engineered out of American life. Before the mid-twentieth century, getting around meant moving your body. The postwar embrace of Euclidean zoning, which rigidly separates residential, commercial, and industrial land uses, made this impossible by design. When your home is miles from the nearest shop, school, or workplace, driving isn’t a choice. It’s a requirement.
Americans now drive an average of 13,662 miles per year — roughly double that of the French, and nearly triple the Japanese. A 2017 Stanford study of 717,527 people across 111 countries found Americans averaged just 4,774 steps per day, ranking 30th out of 46 nations. The decline is generational: in 1969, 48% of children walked or biked to school; by 2009, only 13% did. Meanwhile, occupational physical activity has collapsed. A landmark 2011 study in PLOS ONE found that jobs requiring at least moderate physical activity dropped from 48% in the early 1960s to less than 20% by 2008, representing a loss of roughly 140 calories of daily energy expenditure for men — a figure that closely predicted actual weight gain measured by national surveys.
The health consequences track the infrastructure. By accelerometer measurement (not self-report, which people inflate), only 1 in 20 Americans meets physical activity guidelines.
The comparison with countries that built differently is stark. The Netherlands, where 27% of all trips are made by bicycle across 35,000 kilometers of dedicated cycling paths, has an obesity rate of roughly 15% — one-third America’s 42.7%. Japan, with dense mixed-use neighborhoods and extensive rail networks, sits at 4.5%.
What the fuck are we doing, ya know? We literally built an environment where the most natural thing a human body does — walk — became an act of rebellion.
The food system is rigged at every level
If the built environment removed exercise from daily life, the food system simultaneously flooded it with cheap calories engineered to override biological satiety. This wasn’t accidental. It’s the predictable output of a subsidy structure, a lobbying apparatus, and a food science industry all pulling in the same direction.
The structural tilt begins at the farm. In 2024, corn received $3.2 billion in federal subsidies — 30.5% of all farm support payments. Soybeans got another $1.9 billion. Together, these two crops — the raw materials for high-fructose corn syrup, soybean oil, animal feed, and the building blocks of ultra-processed food — captured nearly half of all federal farm subsidies. Fruits and vegetables, which the USDA recommends filling half your plate with, received less than 1% of direct subsidy payments. Less than 3% of all American cropland grows vegetables, orchards, or berries. The country imports two-thirds of its fresh fruit.
This subsidy architecture makes processed food artificially cheap, and the food industry then engineers it for maximum consumption. Howard Moskowitz, a Harvard-trained psychophysicist, pioneered the concept of the “bliss point” — the precise ratio of sugar, salt, and fat that maximizes craving. Testing 59 variations of sweetness across 3,000 taste tests for a single Dr Pepper product, Moskowitz’s methods became industry standard. Lay’s “Bet you can’t eat just one” wasn’t a joke — it was a product specification.
The consequences became scientifically undeniable in 2019, when NIH researcher Kevin Hall published the first randomized controlled trial of ultra-processed food. Participants eating ultra-processed diets consumed 508 extra calories per day compared to those on unprocessed diets matched for available calories, macronutrients, sugar, sodium, and fiber — and gained nearly a kilogram in two weeks. Those on unprocessed diets lost the same amount. Today, 55% of American calories come from ultra-processed foods. Among children aged 6–11, the figure reaches 64.8%.
The downstream numbers are catastrophic. 40.3% of American adults are obese — up from 15% in 1980. Diet-related chronic diseases cost the healthcare system $334 billion annually in direct medical spending. Meanwhile, 35 to 84 million Americans live in food deserts — areas without convenient access to a supermarket.
The food industry spends nearly $14 billion annually on marketing, more than 80% of it promoting products high in fat, sugar, and salt. When the FTC proposed banning TV advertising to children in 1978, the industry’s lobbying campaign persuaded Congress to strip the FTC of that authority entirely. Let that sink in.
Algorithms harvest attention at the cost of thought
The same optimization logic that governs food engineering now governs information. Social media platforms don’t sell content to users. They sell users’ attention to advertisers. The algorithmic architecture that maximizes this attention reliably amplifies content that provokes outrage, fear, and tribal hostility — because these emotions keep people scrolling.
A 2025 study published in PNAS found that Twitter’s engagement-based algorithm amplified emotionally charged content: 62% of algorithm-selected political tweets expressed anger, compared to 52% in a chronological feed. Critically, users didn’t actually prefer this content — the algorithm overrode their stated preferences to serve what generated more clicks. A separate 2018 MIT study analyzing 126,000 stories found that false news was 70% more likely to be retweeted than true stories and reached audiences six times faster, driven not by bots but by humans’ psychological response to novelty and emotional intensity.
The cognitive environment this creates is measurably degraded. Gloria Mark, a UC Irvine researcher who has tracked attention patterns for two decades, found that the average time a person spends on a single screen before switching dropped from 2.5 minutes in 2004 to 47 seconds today. Americans check their phones 96 times per day. Adults spend an average of seven hours daily on screens — and it takes 25 minutes to recover full attention after an interruption.
Meanwhile, the institutions that once supported informed public discourse are collapsing. More than 2,500 local newspapers have closed since 2005, with papers still shuttering at a rate of two per week.
The cumulative effect is what writer Scott Alexander termed “epistemic learned helplessness” — a state where people, having been repeatedly burned by convincing-sounding but false arguments, rationally give up on evaluating truth claims altogether and retreat into tribal identity or disengagement. Tristan Harris calls it “a race to the bottom of the brain stem.” I’d call it the most profitable cognitive disability in human history.
Loneliness is a design outcome, not a personal failure
The same infrastructure that eliminated walking and flooded the market with engineered food also dismantled the conditions for human connection. Sociologist Ray Oldenburg identified “third places” — pubs, barbershops, community centers, the casual gathering spots that are neither home nor work — as essential to social fabric. His assessment of postwar development was blunt: suburbia was designed to protect people from community rather than connect them to it.
The data confirms this at scale. The U.S. Surgeon General’s 2023 advisory declared loneliness a public health epidemic, reporting that roughly half of American adults experience measurable loneliness. Social disconnection carries the mortality risk of smoking 15 cigarettes per day, exceeds the risk associated with obesity, and increases dementia risk by 50% among older adults.
The “friendship recession” is accelerating. The share of Americans reporting no close friends quadrupled from 3% in 1990 to 12% in 2021, then rose further to 17% by 2024. Those reporting ten or more close friends fell from 33% to 13% over the same period. Time spent with friends dropped from 6.5 hours per week to 4 hours between 2014 and 2019 — before the pandemic made things worse.
Here’s what gets me — we spend 4.9 hours a day on entertainment apps while time with actual humans keeps shrinking. Parasocial relationships with influencers and streamers increasingly substitute for reciprocal social bonds, and the research shows these one-sided connections paradoxically increase loneliness even as they provide a feeling of belonging.
Defaults are never neutral — and yours are set against you
Here we go — this is the through-line that ties it all together.
Behavioral economists Richard Thaler and Cass Sunstein demonstrated a foundational principle: the default option wins. When 401(k) enrollment is opt-in, participation is low. Make it opt-out, and participation reaches 90%. Austria’s opt-out organ donation system achieves 99% consent; Germany’s opt-in system, 12%. The design of the choice, not the preference of the chooser, determines the outcome.
Apply this to daily life and the picture clarifies fast. The “default day” in America — drive alone, eat fast food, scroll social media, stream entertainment — requires zero effort and generates enormous corporate revenue. Meta earned $160 billion in advertising revenue in 2024. The U.S. fast food industry crossed $413 billion. Every one of these industries profits when people consume more and move less.
The healthy alternative — cook from scratch, walk or bike, read deeply, maintain friendships in person — requires active, sustained resistance. Americans spend over $100 billion annually on diets, gym memberships, and supplements. Within two years of starting a diet, fewer than 5% are still adhering. The failure isn’t moral. It’s architectural.
As Dan Buettner, who studied the world’s longest-lived populations, concluded: “Not a single centenarian pursued health. It ensued as a byproduct of where they lived.”
The places that redesigned their defaults got different results
Alright — here’s the part I care most about. Because if this was just a doom story, I wouldn’t have written it. The most powerful evidence that this is structural, not inevitable, comes from the communities that changed the architecture.
Pontevedra, Spain removed cars from its city center in 1999. Today, 70% of trips are made on foot, 80% of children walk to school, and the city has recorded zero pedestrian fatalities since 2011. It gained 12,000 new residents drawn by livability.
Barcelona’s superblocks program, which closes street grids to through-traffic, reduced nitrogen dioxide by 25% in pilot neighborhoods. At full scale — 503 superblocks — health models project 667 premature deaths prevented annually.
Dutch cycling infrastructure, built over decades at roughly €500 million per year, prevents an estimated 6,500 deaths annually and adds half a year to national life expectancy. The health return represents over 3% of GDP on an investment of roughly 0.06% of GDP — a 38-to-1 benefit ratio.
Chile’s 2016 front-of-package warning labels produced a 23% decrease in purchases of high-sugar foods and an 80% consumer awareness rate.
The UK’s tiered sugar levy achieved a 34.3% reduction in sugar sold through soft drinks — while sales volumes actually increased 13.5%, proving commercial viability. Over half of manufacturers reformulated before the levy even took effect. By contrast, the UK’s voluntary sugar reduction program achieved only 3.5%.
The Blue Zones Project, which applies longevity research to American communities through environmental redesign, delivered a 68% reduction in childhood obesity in Redondo Beach schools between 2007 and 2019, with the community achieving the highest well-being score Gallup has ever recorded across 1,500+ communities.
Norway’s school smartphone bans reduced bullying by 43–46% and lowered girls’ mental health-related GP visits by 29%.
Notice what these all have in common. None of them asked individuals to try harder. They changed the environment so that the healthy choice became the easy choice.
The operating principle is environment, not effort
The through-line across every domain — movement, food, attention, social connection — is the same. Environments that make healthy behavior the default produce healthy populations without requiring extraordinary discipline. Environments that make unhealthy behavior the default produce sick, isolated, cognitively fragmented populations regardless of how much those individuals spend on self-improvement.
At the individual level, opting out means recognizing the current is against you and building deliberate counter-structures: choosing a walkable neighborhood over a cheaper suburb, cooking with whole ingredients as routine, imposing hard limits on algorithmic media consumption, and investing disproportionately in face-to-face relationships.
But individual resistance is expensive, exhausting, and inequitable — available primarily to the educated and affluent. The deeper lesson from Barcelona, Pontevedra, the Netherlands, Chile, and the Blue Zones is that structural change is both more effective and more equitable than individual change. The UK sugar levy’s largest benefits accrued to the most deprived children. Chile’s food labels helped low-income consumers most. Norway’s phone bans produced the greatest gains for low-SES girls. When the default changes, everyone benefits — and those with the least capacity for individual resistance benefit the most.
The question isn’t whether people should eat better, move more, think deeper, and connect with others. Nearly everyone already knows this. The question is who sets the defaults — and in whose interest.
Right now, the answer is unambiguous: the defaults are set by industries whose revenue increases when people consume more, move less, and engage passively. Changing this requires not better willpower but better architecture.
The communities that have done it prove it works. The scale of the crisis demands it everywhere else.
Hit me up if you have thoughts and criticisms. I’m always interested in getting better data or having my methodology challenged — that’s how this stuff improves.
For the full source audit with bias analysis, see: Sources & Methodology
For the interactive data journalism version, see: Interactive Edition
Love y’all, hope it serves you.
— Corey
Research assisted by Claude. Data from CDC, PLOS, PNAS, PMC, NIH, USAFacts, Gallup, and cited studies.