The episode opens with back-to-back sponsor reads for Pacific Life Insurance, Hotels.com, and Mint Mobile before Dubner steps in to explain the unusual origin story for this replay. A listener email asked whether this year's southeastern wildfire smoke might have affected students' final exam scores — not because they were displaced, but because the smoke might have impaired their cognition. That question, Dubner says, sent him back to a 2022 episode on exactly that topic, now updated with new facts and figures. It's a quietly clever hook: the listener's real-world curiosity about contemporary wildfires becomes the on-ramp to a rigorous scientific investigation.

With a series of imagined radio announcements — 'the level of particulate matter in the air today is above WHO guidelines' or 'the Supreme Court will delay oral arguments due to high particulate matter in D.C.' — Dubner makes his central thesis visceral: air quality should be as routine a data point as weather or traffic. He introduces particulate matter (PM2.5), tiny particles less than 2.5 micrometers in diameter, invisible to the naked eye, and emitted by everything from dust to fossil fuel combustion. Environmental economist Andrea Linos of Deakin University explains why economists are interested: if pollution impairs cognition and cognition drives productivity, then the true economic costs of dirty air may be far larger than anyone has measured. The stage is set for a detailed scientific investigation into pollution's effects on the human brain.

Greenstone, who co-directs the Climate Impact Lab and spent a year working on climate policy in the Obama White House, introduces the Air Quality Life Index — a satellite-data tool that calculates how much longer people would live if local pollution met WHO standards. The answer is striking: 2.2 years per person on average globally. ^{[2] — Michael Greenstone "Average life lost to pollution: 2.2 years: Michael Greenstone's Air Quality Life Index estimates the average person on Earth lives 2.2 year…" 07:26 Play} The WHO estimates roughly 7 million people die every year from fine particle exposure — more than double COVID deaths in 2020 and more than five times annual car-crash fatalities. ^{[1] — Michael Greenstone "It's worse than cigarette smoking. It's worse than wars. It's worse than auto accidents." 07:40 Play} The economic cost is estimated as high as $6 trillion per year, about 5% of global GDP. Greenstone's framing sets the episode's central tension: we know pollution is catastrophic for physical health, but have we been systematically ignoring its effects on the brain?

Greenstone paints a vivid before-and-after: in the late 1960s, parts of the US looked like Delhi today, with white-collar workers in Gary, Indiana routinely bringing a spare shirt because coal pollution soiled the first. The Clean Air Act of 1970, signed bipartisanly and amended multiple times, changed all that. A 2021 Environmental Research Letters study confirms the downstream benefits: American crop yields are significantly higher thanks to cleaner air, though particulate matter still takes a toll. Today, the US averages just 9 micrograms per cubic meter of PM2.5 annually — above the WHO's 5 µg/m³ threshold, but dramatically better than the China average of 35. Dubner notes the uncomfortable irony: much of the US's clean-air success came by offshoring manufacturing — and its pollution — to developing nations.

In cities of the Western Hemisphere, prevailing winds blow west to east, and east sides are disproportionately poorer — a pattern Hiblik suspected was no coincidence. His team hunted down the locations of 5,000 Victorian industrial chimneys across 70 English cities using historical maps of astonishing detail, then modeled how coal smoke dispersed eastward on the wind. ^{[1] — Stefan Hiblik "As coal smoke darkened English trees, the darker variety of the peppered moth thrived because it could hide from predators. Stefan Hiblik's…" 13:05 Play} To validate the model, they used the peppered moth: the ratio of dark-to-light moths in a given area served as a biological pollution proxy, since darker moths thrived where coal smoke had blackened trees. Census records from 1817 — before heavy coal use — showed no systematic east-west poverty gap. By 1881, at the height of industrialization, east sides across England had a significantly higher share of low-skilled workers. The mechanism: pollution made the east side undesirable, wealthy residents sorted away, and the socioeconomic divide was born.

The most troubling finding from Hiblik's 'East Side Story' research isn't the historical sorting — it's that the sorting has intensified over time. One standard deviation more historical pollution corresponds to a 20% higher share of low-skilled workers in a neighborhood today, compared to 15% in the Victorian era. Hiblik explains this through path dependence: pollution drew the poor east, then highways cut off the east side, school funding fell, building quality declined, and crime rose — each factor reinforcing the last in a classic snowball effect. Test scores in historically high-pollution east sides are lower, and crime rates are higher. The causal question remains open: did pollution impair cognition directly, or did it simply sort lower-income people into those areas? That question is exactly what the next section of the episode sets out to resolve.

Research by economists James Archsmith, Anthony Hayes, and Sudeh Saberian showed that MLB umpires made more wrong calls on high-pollution days — the first strong evidence that air pollution impairs professional adult cognition. But Linos saw its limits: one population, one task. She and her co-author Edson Severnini gained access to Lumosity's Human Cognition Project, pulling data from over 100,000 US users playing games measuring verbal ability, attention, flexibility, working memory, math, processing speed, and problem-solving across more than 4 million gameplay observations over 3 years. ^{[1] — Andrea Linos "Researchers analyzing baseball umpires found they made more wrong calls on high-pollution days. It was the first strong evidence that air p…" 21:41 Play} To handle the fact that pollution isn't measured 'attached to bodies,' the team used wind direction as an instrument: when wind carries pollution from other locations into an area, all residents receive roughly the same dose, creating a natural experiment. The stage is set for a headline result that would shake even the experts.

In one of the episode's most entertaining segments, Dubner convenes a supposedly friendly Lumosity 'Fit Test' with two of his smartest friends — Steve Levitt, economist and Freakonomics co-author, and Angela Duckworth, psychologist and author of Grit. Levitt pre-registers his excuse (18 years of bad sleep), Duckworth pre-registers hers (she's a 'pretty shit sleeper'), and Dubner notes his broken air conditioning. The three play three games: one for mental flexibility, one for memory, and Train of Thought for attention — which Levitt calls 'one of the hardest things I've ever done in my life' while simultaneously scoring in the 97th percentile for his age group. Levitt wins with an average of the 92nd percentile; Duckworth and Dubner tie around the 71st–72nd percentile. Only then does Dubner reveal the experiment's true purpose.

When Dubner reveals that pollution may explain the competition results, Levitt and Duckworth are genuinely surprised. New York had 23.4 micrograms per cubic meter that day; Philadelphia had 24.6; Chicago just 8.7. Linos's model predicts that playing on a high-pollution day shifts your percentile ranking by about 6 points — meaning Dubner and Duckworth, on cleaner-air days, might have been right up there with Levitt. ^{[1] — Steve Levitt "I would say it feels like a 1 in terms of likelihood of being true. And if true, a 10 in terms of importance." 38:10 Play} Levitt's reaction is the episode's funniest: he gives the finding a '1 out of 10' chance of being true, but says if it is true, it deserves a '10 out of 10' for importance. Duckworth, ever the scientist, declines to rate it without reading the paper. The exchange perfectly dramatizes the gap between intuitive skepticism and the weight of the empirical evidence to come.

The episode pauses for three sponsor spots. Southern Company is promoted as an energy company investing $80 billion in infrastructure upgrades to provide reliable and affordable energy. Ozempic — the GLP-1 receptor agonist semaglutide — is promoted as a pill form available in 4 and 9 milligram doses, with listeners directed to a hotline and ozempic.com. TalkAboutPD.com addresses Peyronie's disease, a condition where scar tissue under the penile skin causes painful curvature, encouraging listeners to speak with a urology specialist about non-surgical treatment options.

Acknowledging he's not a medical expert, Severnini synthesizes the biomedical literature: air pollution impairs the brain through two routes. The first is direct — fine particles are small enough to pass through the blood-brain barrier and directly disrupt neuronal function. The second is indirect — particles stimulate the release of pro-inflammatory cytokines, triggering systemic inflammation, oxidative stress, and ultimately neuron loss. Researchers studying this topic consistently observe these same processes: oxidative stress, inflammation, and neuronal damage. Together, these mechanisms explain how short-term variation in pollution exposure — the kind Linos and Severnini measure with daily gameplay data — can translate into measurable shifts in cognitive test scores.

The Linos and Severnini study, published in the Journal of the Association of Environmental and Resource Economists, delivers two major findings. First, the headline: even below EPA and WHO guidelines, air pollution measurably impairs cognition across all seven domains tested — verbal ability, attention, flexibility, working memory, math, processing speed, and problem-solving — for the working-age population, with the largest effects in adults under 50. ^{[1] — Andrea Linos "Even when air pollution stays below EPA and WHO guidelines, cognition is measurably impaired across seven cognitive domains — including mem…" 26:46 Play} Second, and most novel: pollution specifically hits memory. Because memory underlies so much professional knowledge work, the productivity implications are potentially vast. On average, a high-pollution day shifts your cognitive percentile ranking by about 6 points — so Duckworth and Dubner on their high-PM2.5 day in Philly and NYC might have scored as well as Levitt had conditions been equal.

Greenstone has always argued that the 'holy grail' of pollution research is long-run variation — not day-to-day fluctuations, but the kind of persistent difference that policy actually creates. The Huai River heating policy, which gave free coal to provinces north of the river and nothing to provinces south, with limited migration between sides, offered exactly that. ^{[1] — Michael Greenstone "China's Huai River heating policy drew an arbitrary line across the country: free coal for the north, nothing for the south. Because migrat…" 43:50 Play} Among roughly 40,000 people in urban areas within five degrees latitude of the river, people born just north lived about 3 years less than those born just south. In subsequent research, Greenstone extended the analysis to children born between 1975 and 1982: north-side children completed nearly a full year less of education and, observed as adults, earned 13% less. Greenstone calls this the first large-scale evidence on the impacts of long-run early childhood pollution exposure at the concentrations that prevail across much of Asia and sub-Saharan Africa today.

The Huai River cognitive findings lead Greenstone to a sweeping conclusion: if pollution impairs cognition so dramatically, and we haven't been counting that in cost-benefit analyses, we've been understating total losses from dirty air by around 50%. ^{[1] — Michael Greenstone "Understating pollution losses by 50%: Michael Greenstone argues that because we have ignored the cognitive costs of pollution, we have been…" 46:35 Play} That cuts both ways: the benefits of pollution reduction are also 50% larger than we've been claiming, which would justify far more stringent environmental regulations. Greenstone notes that China has achieved in five years what the US never managed after the Clean Air Act: Beijing's PM2.5 fell from over 100 µg/m³ in 2013 to just over 50 by 2018, with Greenstone estimating that a child born in 2018 will live 1.4 years longer than one born in 2013. Meanwhile, the US remains the only G7 country without a coordinated national climate policy — a striking outlier in Greenstone's assessment.

Dubner circles back to the present: the Trump administration's EPA has announced it will no longer weigh the health benefits of reduced pollution when writing clean air regulations. Greenstone's verdict is unsparing — he says it will go down in history books as a dramatic step backward, leading to dirtier air and more greenhouse gas emissions. ^{[1] — Stephen Dubner "The Trump administration announced that the EPA will no longer consider the health benefits of reduced pollution when setting clean air reg…" 49:10 Play} Dubner then offers a personal reflection: he's long wondered whether the climate change debate might have unfolded differently if it had been framed as a fight against pollution rather than a fight against CO2. The evidence for pollution's harms — physiological and now cognitive — is so compelling that it crosses political lines. Naturalists and preservationists have historically come from both parties; no one wants their children breathing dirty air. Severnini gets the final word, urging that particulate matter be part of everyday energy and environmental policy conversations.

Dubner thanks economists Edson Severnini, Andrea Linos, Michael Greenstone, and Stefan Hiblik, as well as Angela Duckworth and Steve Levitt for the brain-game segment, then invites listeners to recommend the show to others. The credits roll — the episode was produced by Zach Lipinski and updated by Dalvin Abuwaji, mixed by Jeremy Johnston, with a full slate of Freakonomics Radio Network staff listed. A short outtake from the Lumosity session plays — Levitt calling it the hardest thing he's ever done, Dubner lamenting that he 'sent those people to their death' in Train of Thought, and Duckworth admitting she feels 'so dumb.' Three closing sponsor reads follow: Made In Cookware (professional-quality cookware trusted in Michelin-starred restaurants), Vitamix (blenders as kitchen multi-tools), and a cross-promo for the new podcast 'A History of the United States in 100 Objects' from 99% Invisible and BBC Studios.