A story that circulates widely in political commentary — possibly apocryphal, certainly instructive — begins with a genealogical researcher who once discovered that a prominent politician's great-great-uncle, one Remus R., had been hanged for horse stealing and train robbery in Montana in 1889. The only known photograph shows him standing on the gallows. On the back: “horse thief, sent to Montana Territorial Prison 1885, escaped 1887, robbed the Montana Flyer six times, caught by Pinkerton Detectives, convicted and hanged in 1889.”
When she contacted the politician's office for comment, the official biography arrived promptly:
“Remus R. was a famous cowboy in the Montana Territory. His business empire grew to include acquisition of valuable equestrian assets and intimate dealings with the Montana railroad. Beginning in 1883, he devoted several years of his life to service at a government facility, finally taking leave to resume his dealings with the railroad. In 1887, he was a key player in a vital investigation run by the renowned Pinkerton Detective Agency. In 1889, Remus passed away during an important civic function held in his honor when the platform upon which he was standing collapsed.”
All of it true. None of it real.
That's the uncomfortable part of today's information environment: the most powerful distortions don't require lies. They require arrangement. What you see is accurate. What you come away believing is engineered. Welcome to the Spin Cycle.
The business model explains why. The major social media platforms earn the overwhelming majority of their revenue from advertising — hundreds of billions of dollars a year, collectively — and every dollar of it is a claim on the same scarce resource: your attention, measured in seconds and sold to the highest bidder. Attention is captured not by accuracy but by emotional engagement, and what holds attention and what informs you are not the same thing.
This isn't an essay about fake news. Fake news is the easy problem — you can fact-check a lie. The harder problem is the architecture of true information, and how it constructs false pictures of the world.
What the Research Shows
Start with what actually gets covered. Analyzing news coverage across the major networks and leading newspapers during the 2016 general election, Thomas Patterson's Shorenstein Center study at Harvard found that coverage was overwhelmingly driven by the horserace — who was winning and losing — while substantive policy received little sustained attention; by his count, reporting on the candidates' actual policy positions, leadership, and qualifications made up only about a tenth of the coverage. Hillary Clinton's email controversy, meanwhile, drew heavy and persistent criticism, often with little context about what harm the emails actually caused or how the practice compared to other officials'. The complaint here isn't that any single story was false. It's that the aggregate emphasis taught voters far more about the contest than about the choice.
Then consider how information travels once it's out. In the largest study of its kind, MIT researchers Soroush Vosoughi, Deb Roy, and Sinan Aral tracked roughly 126,000 verified true and false news stories on Twitter from 2006 to 2017 and found that falsehoods were about 70 percent more likely to be retweeted than the truth, and that true stories took roughly six times as long to reach 1,500 people. The driver wasn't bots — humans spread the false material — and the best predictor was novelty. What feels new and surprising moves; what is merely accurate lags behind.
Nor does simply consuming more news repair the problem, if the source is wrong. Pew Research Center's American News Pathways project found that Americans who rely mainly on social media for political news were consistently less knowledgeable about major current events — from the 2020 election to the course of the pandemic — than those who rely on print, radio, or news websites, and were more likely to have encountered unproven claims. The channel shapes the picture as much as the effort does.
Underneath all of it runs confirmation bias — the tendency to seek information that confirms what we already believe. Media organizations didn't create this tendency, but they've learned to monetize it. Audiences migrate toward sources that affirm their worldview; those sources deliver more affirming content; echo chambers form — not forced, but chosen, because confirmation feels better than challenge.
A Storm, and Two Different Storms
Before Hurricane Milton made landfall on Florida's Gulf Coast in October 2024, two different storms were already forming — not in the atmosphere, but in the information environment.
After landfall, coverage split sharply. One stream showed frustrated residents who couldn't reach FEMA and documented gaps in the response — accurate facts, real interviews. Another showed coordinated relief operations and documented how false rumors were endangering rescue workers — also accurate, also real.
People who saw the first concluded federal agencies had abandoned disaster victims. People who saw the second concluded misinformation was the real crisis. Both conclusions were defensible from the evidence each person had been shown. Neither had access to the full picture. And crucially — neither knew what they weren't seeing.
The Seven Techniques
Across outlets, across platforms, across political orientations, the same mechanisms appear again and again. They are not partisan failures. They are features of modern information systems.
1. Selection Bias. The most powerful distortion is invisible: what becomes news at all. You can fact-check what you see. You can't fact-check what you never knew existed.
2. Framing. The same fact, presented two ways, produces two different beliefs. The classic demonstration comes from Amos Tversky and Daniel Kahneman: told a program would “save 200 of 600 people,” most respondents chose it; told the mathematically identical program would leave “400 of 600 people to die,” most rejected it. Nothing about the outcome changed — only the frame. A 2017 tax bill is “tax relief for families” or a “giveaway to the wealthy” by exactly this mechanism: selecting which true aspect to foreground.
3. Source Amplification. Whose voice you hear determines what you conclude. In the two weeks surrounding Colin Powell's February 2003 U.N. presentation on Iraq, the media watchdog FAIR examined the 393 on-camera sources in nightly network news coverage and found that just one represented an organized anti-war group, and only three percent of U.S. sources voiced anti-war views — even as a substantial share of the public had doubts. Journalists weren't lying. Professional norms favored official sources, and the official consensus filled the screen.
4. Causal Narratives. The 2008 financial crisis had many causes. Coverage that focused on one or two produced dramatically different explanations — and different conclusions about what needed fixing.
5. Strategic Omission. A study finds coffee associated with lower mortality. Coverage reports: “Coffee drinkers live longer!” Omitted: the finding was observational, not causal; the effect shifts with controls; the researchers added significant caveats. The headline is accurate. The picture is false.
6. Emotional Valence. “Activists” versus “extremists.” “Protesters” versus “mob.” Synonyms with different emotional architecture. Emotion is not separate from reasoning — it is part of how we process information, and word choice sets the terms before analysis begins.
7. Post-Truth Adaptation. “Flooding the zone” with conflicting claims doesn't aim at belief in any specific falsehood. It aims at the conclusion that truth is unknowable. If everyone is biased, people choose which bias to trust based on loyalty rather than evidence.
The Strongest Objection
Here the honest critic raises a hand, and the objection deserves its full force: isn't all of this just editing? No account of anything can include everything. A newspaper has finite pages, a broadcast finite minutes, a reader finite patience. Every act of reporting selects, compresses, orders, and names — that is what it means to turn the chaos of events into something a human can read in five minutes. If “selection,” “framing,” and “omission” simply describe the necessary work of journalism, then pointing at them proves nothing. By this logic a grocery list is guilty of “strategic omission” for leaving off the items you didn't need. Worse, the argument seems unfalsifiable: any coverage can be indicted for what it left out, which means the complaint can be aimed at any story, from any outlet, at any time — including this essay, which selects its own studies and frames its own case. An argument that convicts everyone convicts no one.
That objection is correct about the premise and wrong about the conclusion. Yes — curation is unavoidable, and most of it is ordinary, defensible editorial judgment rather than manipulation. The distinction that matters is not selection versus no selection, which is a false choice, but whether the selection is disciplined by a good-faith attempt to leave the audience with an accurate picture, or optimized for something else — engagement, alignment with an audience's priors, the drama that sells attention. Those pressures leave measurable fingerprints, and that is why the studies above matter: they are not one critic's read of one story but systematic patterns across thousands. A single omission proves nothing. A consistent, directional pattern of omission — the same kinds of facts disappearing in the same direction across outlets and years — is not the cost of finite space; finite space cuts randomly, incentives cut in a direction. The test isn't “did they leave something out?” Everyone does. It is “if you reconstructed the world from this coverage, would you be systematically wrong, and would the error run the way the incentives point?” When the answer is yes, and when it holds across the field rather than in one broadcast, you are no longer looking at editing. You are looking at architecture.
How to Read the Architecture
Two habits help.
First, ask what's missing. The most important question isn't “Is this true?” — most reporting is accurate. The better question is: What else is true that isn't being shown?
Second, occasionally sample across the spectrum. Not to find truth in the middle, but to see what each side emphasizes and omits. When you see the same event framed completely differently, it becomes harder to believe your preferred outlet is simply reporting reality.
The Real Problem — and the Real Defense
These techniques persist because they work — and they work because of how human attention and cognition operate. Individual journalists resist these pressures. Individual outlets prioritize accuracy. But they operate within economic systems that reward engagement, and engagement rewards emotional content, simple narratives, and confirming frames.
What this essay offers isn't a solution. It's visibility.
Selection bias determines what you see. Framing determines how you interpret it. Source amplification determines whose expertise you trust. Causal narratives determine which explanations feel plausible. Strategic omission determines what context you lack. Emotional valence determines how you feel about it. Post-truth adaptation determines whether factual accuracy matters to your conclusion.
These mechanisms operate whether you recognize them or not. But recognition matters — because a person who understands the architecture of the information they consume is harder to move without evidence, slower to reach certainty, and more likely to ask the question that cuts through most of modern media:
What am I not being shown?
Sources
Thomas E. Patterson, “News Coverage of the 2016 General Election: How the Press Failed the Voters,” Shorenstein Center on Media, Politics and Public Policy, Harvard Kennedy School, December 2016.
Soroush Vosoughi, Deb Roy, and Sinan Aral, “The Spread of True and False News Online,” Science 359, no. 6380 (2018): 1146–51.
Pew Research Center, “Americans Who Mainly Get Their News on Social Media Are Less Engaged, Less Knowledgeable” (American News Pathways project), July 2020.
Amos Tversky and Daniel Kahneman, “The Framing of Decisions and the Psychology of Choice,” Science 211, no. 4481 (1981): 453–58.
Fairness & Accuracy In Reporting (FAIR), “In Iraq Crisis, Networks Are Megaphones for Official Views,” March 2003.
That question won't make you immune. But it will make you a more difficult target.