LeoNes

How We Know What's True

The world's best lie-detector How We Know Anything · The Testable Guess · Fair Tests · The Great Trap · Checking Each Other · Spotting Fakes. Science isn't a pile of facts — it's a method for being less wrong on purpose, the only system humans ever built that admits its own mistakes and fixes them.

Ages 14–16

What's inside

• How We Know Anything — evidence beats opinion

  Every day the world hands you thousands of claims — an ad, a headline, a friend, a video, a gut feeling. Some are true, many aren't, and none come with a label. The single most useful mental habit you can build is to meet each one with a quiet question: how do you know? Science is, at its heart, just the most disciplined possible answer to that question — a way of deciding what to believe based on evidence rather than on who's shouting, who's famous, or what feels right.

• The Testable Guess — hypothesis and prediction

  Curiosity starts with a question — why does bread go mouldy faster in the bin than the fridge? To turn that into science, you sharpen it into a hypothesis: a specific, testable guess at the answer ('mould grows faster in warmth'). The crucial word is testable. A good hypothesis doesn't just explain what you've already seen — it sticks its neck out and predicts something new: 'so bread kept warm should go mouldy sooner than identical bread kept cold'. Now you can actually check.

• Fair Tests — controls, variables, sample size

  Having a testable prediction is only half the job; now you have to test it fairly, and that's harder than it looks. Say you want to know if a new plant food really works. If you feed it to one plant on a sunny windowsill and compare it to a starved plant in a dark cupboard, and the fed one thrives, you've learned nothing — was it the food, or the sunlight, or the warmth? A fair test changes only the one thing you're studying and holds everything else identical.

• The Great Trap — correlation is not causation

  Here's the single most useful trap to learn to dodge, because you'll meet it every day. When two things rise and fall together — they 'correlate' — it is desperately tempting to assume one is causing the other. Sometimes it is. Very often it isn't. Ice cream sales and drownings climb at the same time of year; a tabloid could scream 'ICE CREAM CAUSES DROWNING'. But of course neither causes the other — a hidden third thing, hot summer weather, drives both. Spotting that hidden third factor is one of the sharpest mental moves you can make.

• Checking Each Other — peer review and replication

  A single clever person, however brilliant, can be fooled — by their own hopes, a hidden mistake, or plain wishful thinking. So science doesn't trust individuals; it trusts a process built on relentless mutual checking. Before a study is published, it's sent to other experts in the same field — the scientist's 'peers' — whose job is to hunt for flaws, holes and overreach, and who can send it back or reject it outright. This peer review is the field's first, imperfect filter, and it's the reason a claim 'published in a scientific journal' carries more weight than the same claim shouted on a podcast.

• Spotting Fakes — how to smell pseudoscience

  You now have the whole toolkit: evidence beats opinion, real ideas are testable, tests must be fair, correlation isn't cause, and no result counts until others confirm it. The final skill is putting it all to work as a detector — because a huge industry exists to sell you things that wear the costume of science while quietly ditching its substance. Pseudoscience is the polished imitation: the confident language, the graphs, the 'clinically proven', the person in a white coat — with the actual testing surgically removed.

Keep exploring

Other languages

Loading LeoNes…