A ten-slide series on the second move in the playbook: repeat a claim until it feels like something you already know. This page has the studies, the transcripts, the counts, and the record behind every card, with dates, links, and what each one cannot tell you, plus a reading check you can try yourself.
The full text of every card, in order, with the card itself under each one. Select a card to open it full size.
Grouped by card. Each source has a summary of what it says, a scope line saying how far it reaches, and a link to the original.
Two experiments with 40 Duke University undergraduates each. Students first read a set of statements and rated how interesting each one was, then rated a larger set for truth on a six-point scale, then took a knowledge check. False statements they had read earlier were rated as more likely to be true than new ones, including statements that contradicted answers the same students got right on the knowledge check. One example in the paper is a statement that a sari is the short pleated skirt worn by Scots.
Seven studies with a combined 2,196 participants. Every study showed the effect, and the authors found no evidence that any of the cognitive measures they tested changed its size. Those measures included intelligence tests, the need for quick and firm answers, and a preference for intuitive or deliberate thinking.
Three online studies using real false and true news headlines shown in a social media format. Seeing a false headline once raised its later accuracy rating: 1.77 against 1.65 on a four-point scale in the second study, with 949 participants. The effect held for headlines that ran against the reader’s politics and was still present one week later.
Two experiments showed statements between 1 and 27 times, then asked for truth ratings a week later. Ratings kept rising with more repetitions, but the largest jump came from the second encounter, with smaller gains after that.
The study usually credited with first measuring the effect. Participants rated how sure they were that each of 60 statements was true or false across several sessions, and rated validity rose for the statements that repeated.
A synthesis of 182 studies and 31,184 participants, published between 1977 and 2025. After correcting for the tendency of small studies to report larger effects, the pooled effect was g = 0.37, which the authors describe as small, with results varying a good deal from study to study.
Section 4: “I am directing that it officially be renamed the Gulf of America.” The order directs guidance to ensure all federal references use the new name, including agency maps, contracts, documents, and communications. A proclamation declaring “Gulf of America Day” followed on February 9, 2025.
January 20 to December 31, 2025: 73 documents used only “Gulf of America,” 62 used both names, and 44 used only “Gulf of Mexico.” January 1 to October 14, 2026: 116 used only the new name, 29 used both, and 17 used only the old one. For comparison, “Gulf of Mexico” appeared in 262 documents in the same months of 2024.
The White House limited the Associated Press’s access after it kept using “Gulf of Mexico.” A deputy chief of staff wrote that the AP “continues to ignore the lawful geographic name change.” The judge granted the AP a preliminary injunction, writing that the government cannot shut doors to journalists “because of their viewpoints.”
The order authorizes “Department of War” as an “additional secondary title” for the Department of Defense. The name appeared in 14 Federal Register documents in 2025 and 230 in 2026 through October 14. Military Times reported in June 2026 that the legal name remains the Department of Defense until Congress acts.
Factba.se publishes searchable transcripts of the president’s speeches, interviews, and remarks. For each claim, my research assistant counted distinct events, January 20, 2025 to October 14, 2026, where he said the key words in a passage on that topic. The prisons claim appeared at least 186 times (66 in 2026). The rigged-election claim, at least 64 times (30 in 2026). The investment figure, at least 169 times (96 in 2026).
PolitiFact rated the claim that countries are emptying their prisons into the United States Pants on Fire: “There is no evidence this is happening.” CNN’s Daniel Dale listed it among the president’s most repeated false claims of 2025 and reported that no proof was ever produced. Recent examples, from transcripts: June 10, 2026, “jails and mental institutions from all over the world were emptied out”; June 23, 2026, “Some of the countries had no prison population left.”
The joint statement of the federal and state election infrastructure councils, published by the president’s own election security agency in his first term, began: “The November 3rd election was the most secure in American history.” Biden won 306 electoral votes to 232. In 2026 the FBI took more than 650 boxes of 2020 material from Fulton County, Georgia; NPR’s review of the state’s files found one of the search affidavit’s five claims partially substantiated, and a 2024 state report said the problems found did not affect the accuracy of the results.
The figure he gives has climbed from $17 trillion in 2025 to $21 and $22 trillion in September 2026. The White House’s own investments page lists 277 commitments; added up, they come to about $11.25 trillion. PolitiFact rated “$22 trillion coming into this country” False and noted that even the White House list includes multi-year goals and pledges that may not happen.
After the shooting at the White House Correspondents’ Dinner on April 25, 2026, Snopes documented 16 accounts posting essentially the same message about building the White House ballroom within about 90 minutes, from 9:28 to 10:55 p.m. Eastern. The wording was similar, not identical, for example “That White House Ballroom Is Looking Better and Better” and “BUILD the WHITE HOUSE BALLROOM NOW!”
The Times tracked at least 304 TikTok accounts with AI-generated faces, all using the same caption professing love for God, America, and Trump, plus more found by researchers on other platforms. Experts could not say who was behind them. The Republican National Committee said it was not involved.
Economists at the New York Fed estimated the share of the 2025 tariffs borne by U.S. firms and consumers: 94 percent from January to August, 92 percent in September and October, and 86 percent in November. Foreign exporters bore 6, 8, and 14 percent. Their conclusion: “U.S. firms and consumers continue to bear the bulk of the economic burden.”
In the State of the Union on February 24, 2026, the president described “the tariffs, paid for by foreign countries.” FactCheck.org reported that many economists dispute this and cited the New York Fed work. It also found “no evidence of a broad decrease in U.S. drug prices.” The same claim about tariffs appears in transcripts and posts from February 2025 through September 2026, and in an op-ed published under his name on January 30, 2026.
The index for prescription drugs was 572.655 in January 2025 and 543.556 in September 2026, a fall of 5.08 percent. The president has said drug prices are being cut by “1,000 percent” or by “400, 500, 600 percent” in at least 47 events since July 2025. Since January 2026 he has said there are “two ways of calculating.” His own example in May 2026, a drug that went from $966 to $168, is a cut of 82.6 percent; the old price was 475 percent higher than the new one.
At a White House event on September 22, 2025, the president told pregnant women “Don’t take Tylenol” around a dozen times, AP reported. Experts called the link unproven. The press secretary said the administration does not believe “popping more pills is always the answer.” The same all-caps post, word for word, was published or reposted on his account four times between September 26, 2025 and January 5, 2026.
The notice asks clinicians to consider minimizing acetaminophen for routine low-grade fevers in pregnancy. It also says “a causal relationship has not been established,” that “there are contrary studies in the scientific literature,” and that acetaminophen “is the safest over-the-counter alternative in pregnancy.”
2,480,797 children born in Sweden from 1995 to 2019, of whom 185,909 were exposed to acetaminophen before birth. Without comparing siblings, there was a slight association with autism (hazard ratio 1.05). Comparing siblings, which accounts for shared family factors, there was none (0.98). The authors concluded that earlier associations may have come from family factors, not the drug.
The CDC’s page now opens by calling the statement “vaccines do not cause autism” not evidence-based. A study of 657,461 Danish children found no increased autism risk for children who received the MMR vaccine (hazard ratio 0.93).
Four experiments with 359 participants. One group rated each statement for truth the first time they saw it; the other rated how interesting it was. Later, the group that had rated for truth showed no repetition effect for falsehoods they had the knowledge to catch.
The handbook recommends building a correction in up to four parts: state the fact, warn that a myth is coming and mention it once, explain how the myth misleads, and end by restating the fact.
Three experiments with 1,718 participants tested whether a correction that introduces people to a false claim leaves them more misinformed a week later than people who saw nothing. Overall it did not.
Read the article below, then try the six questions. Each one uses a different reading skill from part two, in rough order of difficulty. Write your answer down before you open the strong answer.
Paragraph 14 says many of the new tariffs “will be stacked on top of one another.” What does stacked mean here?
Several different tariffs apply to the same goods at once, so their rates add together. Paragraph 15 shows it: 10% plus 10% plus a 3% base rate makes 23%.
Paragraphs 14 and 15
According to the Dallas Federal Reserve, what would the Fed’s preferred inflation measure have been in March without tariffs, and what was it actually?
About 2.3% at an annual rate without tariffs, compared with an actual 3.2%.
Paragraph 9
In one or two sentences of your own, what is the author’s main argument?
The new tariffs announced in July are built to last and are added on top of older ones, so the cost to U.S. consumers will keep growing. The claim that foreigners pay is wrong, because the bill goes to the U.S. businesses that import the goods.
Paragraphs 4, 5, 7 and 8
Paragraph 8 says tariff bills go to U.S. businesses that import the goods, and paragraph 2 says the Supreme Court overturned the emergency tariffs. If those tariffs were refunded, who would you expect to get the money back?
The U.S. importers who paid the bills, not the shoppers who paid higher prices. The article never says this; you have to put paragraphs 2 and 8 together. Card 6 is about the same question of who pays.
Paragraphs 2 and 8
How are the new July tariffs different from the 2025 “Liberation Day” tariffs, according to the author?
They are similar in scope and impact, but they rest on existing trade law, which the administration sees as beyond the reach of the Supreme Court’s review, and most are designed to be permanent rather than temporary.
Paragraphs 3, 4 and 18
Paragraph 8 says tariff invoices are “issued by the Treasury Department.” How could you check that, and what would you find?
Check U.S. Customs and Border Protection at cbp.gov. CBP, part of Homeland Security, assesses and bills the duties; the money then goes to the Treasury. The author’s main point still holds, since the bill goes to the U.S. importer, but this detail is off. Also notice the sentences that begin “I believe” in paragraphs 4 and 21: those are his judgments, not reported facts.
Paragraph 8, with paragraphs 4 and 21
Connect it to the cards: paragraph 7 calls the claim that foreigners pay the tariffs erroneous, which is the claim on card 6. Download the printable worksheet (PDF), with room to write and an answer key. Want more practice? Play Between the Lines: read a current news article, answer twelve questions, and see your reading level.
The joke move at least has a speaker you can look at. This one does not. No single repetition is the problem. The claim only has to be around, again and again, until it feels like something you have always known.
What stayed with me from the research is that knowing better is not the protection I assumed. Students who could give the right answer still rated the wrong one a little higher after seeing it once. That is why I counted. When a claim has been said 186 times, the 187th time does not feel like a claim anymore.
I chose examples from the current administration because they are the ones my kids are hearing, and because the transcripts make the repetition countable. The same habit is worth having with anyone: when something feels like common knowledge, ask whether you know it or have only heard it.