Your brain lies to you.
The last room showed your eyes guessing; this one shows your mind doing it. Five mental shortcuts everyone runs — including me. They’re not stupidity and they’re not sin; a bias is just a shortcut that’s usually helpful and occasionally wrong. Knowing their names is how you catch them in yourself. Disagreement isn’t a moral failure — it’s often just two people running different shortcuts.
1. Anchoring — the first number sticks
Whatever number you hear first quietly drags every number after it. Ask people if the tallest tree is taller or shorter than 1,000 feet, then ask them to guess its height, and they’ll guess higher than a group who first heard “180 feet” — same tree.
The check: decide what a thing is worth before you look at the asking number. Bring your own anchor.
2. Confirmation bias — we shop for agreement
We notice, remember, and believe the things that fit what we already think, and quietly skip the rest. It feels like “doing research”; often it’s collecting a scrapbook of yeses.
The check: go looking for the best version of the other side on purpose — the devil’s advocate. If you can’t state it fairly, you don’t understand it yet.
3. The halo effect — one good trait paints the rest
If someone is attractive, confident, or well-spoken, we assume they’re also smarter, kinder, and more honest — with no evidence for the extras. One bright trait throws a halo over all the others.
The check: score the claim, not the charisma. Ask “what’s the actual evidence?” separately from “do I like this person?”
4. The availability trap — vivid feels common
The easier something is to picture, the more likely it feels. Dramatic, recent, or scary events come to mind fast, so we overrate them; quiet, common risks stay invisible.
The check: ask “how often does this really happen?” and look for the base rate — the boring number behind the vivid story.
5. Dunning–Kruger — the less you know, the surer you feel
When you’re new at something, you can’t yet see how much you’re missing — so confidence runs ahead of skill. The cure is the same skill that reveals the gap: the more you actually learn, the more you notice what you don’t know.
The check: when you feel most certain, ask what would prove you wrong — and whether you’ve honestly gone looking.
A bias isn’t a flaw in the human; it’s a feature that saves time and sometimes costs truth. Naming it is how you keep the discount without paying the price.
The machine has every bias too, and no idea it has them. Only a human can name the trap and step around it. ;