What a four-month experiment with 13,500 Bihar schoolchildren reveals about how to teach children to spot fake news, and why the lesson starts with the adults who forward.
At twenty to eleven on a weeknight, a circular lands in the Class 10 parents’ group. It carries the board’s emblem, a reference number and a signature, and it says the Science paper has been postponed by a week. Within minutes, Anika’s mother has forwarded it to the family group with a single line of relief, and Anika closes her textbook for the night. By morning, the board has called the circular a forgery, and the paper is still four days away. If you have ever wondered how to teach children to spot fake news, this small household scene holds most of the answer, and most of that answer concerns the adults.
Anika and her mother are a composite rather than a family I know. I have drawn them from a pattern that repeats in a great many homes every examination season. An article about checking sources ought to tell you plainly which of its characters you cannot look up. Everything else that follows, every study, circular and number, is real and linked.
Notice who forwarded the circular. It was not the teenager glued to her phone, whom we usually picture as the weak link. Instead, it was a careful, loving parent at the end of a long day. That matters, because children rarely learn to doubt a forward from a lecture. They learn it, or fail to, by watching what the adults around them do at twenty to eleven.
Why sensible, loving people forward false things
Their attention was somewhere else
Start with some reassurance, because the research offers it. Researchers Gordon Pennycook, David Rand and their colleagues ran a set of experiments, published in Nature. They found that whether a headline was true barely changed whether people wanted to share it, even though it strongly changed whether they believed it.
Most of the same people said that sharing only accurate news mattered to them. So they were not choosing falsehood. Their attention had simply drifted to other questions: will this help someone, will the group find it interesting, will I look well informed?
When the researchers nudged attention back to accuracy, the quality of what people shared went up. In one survey experiment, asking people to consider accuracy first cut the sharing of false headlines in half. The forwarder, in short, is rarely foolish. Usually, they are busy, kind and thinking about something else.
Feeling first, checking later
Emotion widens that gap. A separate set of studies involved more than four thousand people. In them, leaning on feelings rather than reasoning made people more likely to believe false news, while their belief in true news stayed the same. Fear and relief are especially strong pulls, and an examination season supplies both in bulk.
Anika’s mother did not stop to check, because the circular promised exactly what she wanted for her daughter. Children feel the same pull. A message warning of a leaked paper, or one ranking the class before results are out, travels on the same current. The class WhatsApp group already carries plenty of comparison and worry, as I wrote in the piece on anxiety in children, and every unchecked forward adds to that load.
The forgery that leaned on a real event
The most convincing fakes rarely start from nothing. In May 2026, CBSE’s portal for answer-book photocopies struggled under heavy traffic, and the board extended its deadline. The very next day, a forged circular claimed the entire re-evaluation process had been cancelled. It falsely carried the name of the board’s Controller of Examinations. Worse, it arrived just when anxious families expected bad news. The board answered on its official account, called the notice fake, and asked everyone to verify information on its own website. Earlier that year, another forgery claimed board papers in the Middle East had been cancelled. Each time, a genuine disruption made a false message feel plausible.
A chip in every note
Not every forward does harm, and a few are worth a smile. In November 2016, within hours of the new ₹2,000 note being announced, forwards claimed that each note carried a nano GPS chip which satellites could track. The Reserve Bank’s spokesperson replied that no such technology existed anywhere in the world. Years later, when the note was withdrawn, Kerala Tourism posted a plate of banana chips with a caption promising that these, at least, would stay in circulation. The episode is funny now. Even so, it shows the whole pattern in miniature: a confident technical claim, an exciting moment, and nobody pausing to ask how a slip of paper could talk to a satellite.
Why a tips sheet is not how to teach children to spot fake news
A gain that faded within weeks
Here is the part that surprised me most. The obvious response to misinformation is a list of tips: check the source, look at the date, distrust shouting capital letters. Such a list had already been shown to people in fourteen countries, in what researchers describe as the world’s largest media literacy campaign.
A team led by political scientist Andrew Guess tested a version of it. In their study, the tips helped a highly educated online group in India tell mainstream news from false news 17.5 per cent better. That was a real gain. Several weeks later, however, the researchers could no longer detect the improvement in the Indian group. Among a representative group in a largely rural part of northern India, the tips made no difference at all.
One hour, in an election season
A second Indian study pushed further. During the 2019 general election, political scientist Sumitra Badrinathan gave 1,224 people in Bihar an hour-long, face-to-face session on spotting misinformation. On average, the hour did not make them any better at identifying false stories. Belief proved stickier than skill, particularly where a false story agreed with what people already wanted to believe. Taken together, the two studies point the same way. A single dose of good advice rarely changes what people do with the next forward.
Nearly twenty lakh children, two and a half hours
Kerala went bigger than anyone. In 2022, its education technology agency, KITE, trained 19.72 lakh students from Classes 5 to 10 to recognise and stop fake news. The state has since added fake-news detection to its school ICT textbooks. That ambition deserves credit. Still, the training itself ran for two and a half hours, and I could find no published measure of what it changed. One thread in Kerala’s work looks more promising, though. Its Little KITEs programme includes a strand in which students teach their mothers digital skills, including how to avoid fake news and scams. Hold on to that picture of the child as teacher. It returns in the strongest study of all.
What 13,500 children in Bihar learned to do
Four lessons spread across fourteen weeks
In late 2023, a team of four researchers, Priyadarshi Amar, Sumitra Badrinathan, Simon Chauchard and Florian Sichart, tried something slower. Working with Jeevika, the Bihar government’s rural livelihoods agency, they offered a free certificate course to more than 13,500 students aged 13 to 18 across 583 villages.
Students in half the villages attended four 90-minute sessions on health misinformation, spaced roughly three weeks apart over fourteen weeks. The other half took four lessons in spoken English, so both groups gave up the same evenings after school. Between sessions, the misinformation group took home short summary sheets and homework: stories to write, things to observe, and conversations to have with their families. The results appeared in the American Political Science Review in October 2025.
Habits of mind rather than a checklist
What the course taught mattered as much as how long it ran. Deliberately, its designers chose thinking over prescriptive tips. Students learned to notice when a message was working on their emotions, and to stop treating a messenger’s shared background as proof that the messenger was right. They also learned to ask who would genuinely know about a subject; for health, that meant the local ASHA worker rather than whoever happened to forward the claim. Students practised pausing before passing anything on, too. In one classroom exercise, a student played a parent who had repeated a health myth at dinner, while another, playing the child, tried to respond without giving offence.
The lessons lasted, and travelled
By the end, students who took the course had become noticeably better at telling true health stories from false ones, and better at deciding what was worth sharing. Crucially, six of the eight test stories had never come up in class, so the students had to apply what they had learned to fresh material. The researchers estimate that the course moved a typical student from the middle of the untrained group to ahead of about six in ten of them.
Four months later, they surveyed a random group of roughly two thousand students again, and the advantage had held. It had also spread to a subject the course never touched. India held a general election in between, and the trained students proved better at judging viral political stories they had never seen before.
Then the parents improved too
The finding that stopped me came last. At the four-month follow-up, the researchers also interviewed one parent or guardian in each household. Parents of trained students had become better at spotting false health information themselves. Nobody had taught those parents anything. Instead, the skill appears to have travelled up through the family, and the authors suggest that the homework and take-home sheets may have carried it. In a country where children are raised to defer to their elders, that is no small thing. Given time and a little structure, a child became the household’s quiet fact-checker.
A fair warning about doubt
One result deserves an honest mention. Straight after the course, trained students were slightly more sceptical of all news, true stories included. The researchers judged this a reasonable trade, since their students began by trusting almost everything. More reassuringly, by the four-month follow-up the extra doubt about true stories had disappeared, while the sharper judgement about false ones remained. For a parent, the lesson is plain. We want a child who weighs a claim, not one who waves every claim away.
What the study does not promise
The authors are careful, and we should be too. Their course was a well-run programme with trained teachers, government backing and a certificate at the end. They caution against assuming the same results would follow anywhere else. Most of these students did not even own a phone; only about one in ten did.
Nobody should read the study as proof that a parent can reproduce a field experiment at the dining table. What it does show is the shape of what works: repetition, discussion, real examples, and learning carried home. For any parent asking how to teach children to spot fake news, that shape is the useful part, because a family can offer every piece of it, free, for years.
Thrive Tip
Choose one forward a week, from any group, and bring it to dinner. Do not say whether you think it is true. Ask your child to decide, and to tell you how they decided.
Then hand them your own phone and let them check yours.
Critical thinking, in the words a family actually uses
Three questions that do most of the work
Critical thinking sounds like a school subject, yet inside a family it comes down to three questions. Who is telling me this? How would they know? And who else, with no reason to agree, says the same thing? A claim that survives all three has earned some trust. One that fails the first two deserves a pause, however urgent it feels. For a forward, the three questions are enough.
What those questions sound like at home
Definitions only go so far, so here is how the questions sound in an ordinary house. A message announces that schools will be shut tomorrow for heavy rain, and a twelve-year-old asks, “Has the school or the collector said so?” A forward claims a famous scientist has predicted a heatwave hot enough to cook eggs on the road, and a teenager asks, “Which scientist, and where did they actually say it?” Someone shares a screenshot of a “leaked” board paper, and the question becomes, “If it really leaked, why would a family group have it before the board has said a word?”
Notice that none of these replies calls the forward false. Each one simply asks how the sender could know. That is the whole posture, and children much younger than our teenagers can learn it.
How to teach children to spot fake news at home: seven habits
Each of the seven habits below costs nothing, and none needs a lesson plan. They work at any age with small adjustments. Children of ten or eleven do best with claims close to home, such as an advertisement promising that a drink will make them taller. Teenagers, meanwhile, can take on the real forwards in family and class groups. Whatever the age, keep the tone curious rather than suspicious. The goal is a child who enjoys finding out.
Make the pause a family rule
Begin with the cheapest habit of all. Since inattention drives so much sharing, a short stop before forwarding does more than any trick for spotting fakes. Make the rule explicit and make it shared: in this house, nothing goes out until we have read it twice.
Pair the pause with one slow breath. A review of fifteen small studies on healthy adults linked slow breathing, the unhurried kind many pranayama practices use, with feeling calmer and less stirred up. The evidence is modest, but a breath costs nothing and gives reason a moment to catch up with feeling. Late-night forwards, meanwhile, can wait for morning. A teenager’s sleep is worth more than any group’s replies, a case I make at length in the article on teenage sleep.
Name the feeling before judging the facts
Next, teach your child to notice what a message is doing to them. Is it making them frightened, furious, relieved or pleased with themselves? A strong feeling does not prove a claim false, but it is a signal to slow down, because that is exactly when judgement slips. You can model this aloud: “This one is making me anxious, so I am going to check it before I believe it.” Children who hear an adult name a feeling learn that a feeling can be noticed rather than obeyed.
Ask who would actually know
The Bihar course taught students to match each claim to the authority that would genuinely know, and that single move settles most family forwards. A circular about board examinations gets checked on the board’s own website or official account, which is precisely what CBSE asks families to do. The school or the district collector confirms a rain holiday. Anything about banknotes belongs to the Reserve Bank. Health claims go to your family doctor or the local health worker, never simply to whoever sent them. Teach it as one question your child can ask about anything: who owns this information? Usually, the answer is a single search away.
Read sideways, not deeper
When a forward links to a website, the instinct is to study that page closely. Professional fact-checkers do the opposite. In a Stanford study, education researchers Sam Wineburg and Sarah McGrew watched historians, undergraduates and fact-checkers judge unfamiliar websites. Historians and students stayed on the page and were taken in by official-looking logos and web addresses. Fact-checkers, by contrast, left quickly, opened new tabs, and checked what others said about the site. As a result, the fact-checkers reached sounder conclusions in a fraction of the time. The researchers call this reading laterally, which simply means reading sideways. For a child, it becomes a rule: before you trust a page, find out what other people say about it.
Learn what the double arrow means
WhatsApp itself offers a clue that most families never notice. Once a message has passed through a chain of five or more chats, it carries the label “Forwarded many times” beside a double arrow. From then on, you can send it to only one chat at a time. The label does not mean the message is false. It does mean that nobody in your group wrote it, and that its first sender is probably a stranger. So show your child the double arrow on a real message. Then agree together that a double arrow always triggers the three questions.
Say “I don’t know yet” out loud
A child who believes adults always know will never learn to check. When a forward stumps you, say the honest words: “I don’t know yet. Let’s find out.” Then find out together, while your child watches. I have argued before, in the essay on what “educate” really means, that this small admission teaches more than any answer. With forwards, it does one thing more. It shows a child that careful, kind people check too.
Let your child check your forwards
Finally, reverse the roles. Before you send something to a family group, ask your teenager to check it first, and accept the verdict gracefully. The Bihar course rehearsed exactly this scene, a child answering a parent’s dinner-table myth, because it is the hardest conversation for an Indian child to begin. When you invite it, the conversation stops feeling disrespectful and becomes a service. Over time, you may find, as the Bihar parents did, that your own judgement sharpens alongside your child’s.
Thrive Tip
Agree on one short family sentence for unchecked forwards, such as “Is this checked?”
Anyone in the family, of any age, may ask it of anyone else, and nobody is allowed to take offence.
When the child must question an elder
Why correcting is harder than doubting
The Bihar researchers found something every Indian parent will recognise. Trained students became more willing to hold back from sharing false stories, yet stayed hesitant to correct the people who shared them.
The authors tie this to norms that discourage confronting others. Girls, in particular, grew no more willing to act publicly, although they improved at judging stories just as much as boys did. Deference to elders is part of what makes a household warm, and no child should be asked to give it up. The task is to give children a respectful way to raise a doubt, so that respect and accuracy stop pulling in opposite directions. If you have a daughter, tell her in so many words that she is allowed to do this.
Question the message, never the person
Give your child words that doubt the forward while honouring the sender. “Grandma, this looks important. Shall we check where it came from before we send it on?” works far better than “This is fake.” So does “The board’s website says something different. Can I show you?” Sentences like these invite an elder to check alongside the child. By contrast, a flat verdict, however correct, asks that elder to lose face in front of the whole family.
Stand beside your child when they speak up
Children will only question an elder if they trust that a parent will stand beside them. So when your child queries a forward in the family group, thank them in the group, not only in private. If the forward came from you, say so cheerfully and delete it. A teenager who watches a parent take a correction well learns more from that moment than from any lesson.
Three ways parents undo the lesson
Mocking whoever sent it
It is tempting to laugh at the relative who believed in the GPS chip. The claim deserves the laugh; the relative does not, at least not in front of the children. Mockery teaches that being fooled is shameful, and a child who fears shame will hide mistakes rather than check them. Besides, the research on attention suggests that the relative was probably distracted rather than gullible, which could describe any of us on a busy evening.
Teaching that nothing can be trusted
Alarmed parents sometimes swing to the opposite pole and tell children that everything online is lies. That is not critical thinking. Rather, it is another way of not thinking, and it leaves a child just as easy to mislead. Someone who trusts nothing can be persuaded that the real warning is the fake one. Aim instead for a child who can explain why one source deserves more trust than another.
Turning a habit into a lecture
Lastly, the habit dies when every forward becomes a sermon. The Bihar course worked through discussion, examples and practice spread over months. Children who are only told what to believe learn obedience instead, a pattern I explored in the essay on the obedient classroom. So keep it light, keep it regular, and let your child win often. Ten minutes over dinner each week will do more than one long talk a year.
Back to the circular at twenty to eleven
Return, for a moment, to Anika’s house, and imagine the same kind of circular arriving a year later. By then the family has spent a year bringing forwards to dinner, and everyone knows the family sentence. This time, Anika sees it first. She notices her own relief, which she has learned to treat as a reason to slow down. Then she looks for a matching notice on the board’s website, finds none, and types three words into the family group: “Is this checked?” Her mother replies with a laughing face and deletes her own forward.
Nothing dramatic has happened. The Science paper is still four days away, and Anika is still revising. Yet something durable has taken root in that house, and it will outlast every board examination she sits. The children in Bihar needed fourteen weeks and a certificate course to get there. Your child has something the course could only approximate: a family that checks together, at the dining table, for years. That, in the end, is how to teach children to spot fake news. It begins with whoever forwards next.
