Why biases exist at all
A bias is a shortcut's shadow. The mind's quick answers (what psychologists since Tversky and Kahneman call heuristics) work often enough to keep; the biases are their systematic misses. You can't uninstall them. The realistic goal is knowing the conditions under which your own answers deserve suspicion.
1. Confirmation bias
You search for, notice, and remember what agrees with you. It's why two people leave the same meeting with opposite victories. The working fix is procedural, not willpower: ask what evidence would change your mind before looking, then look for that.
2. Anchoring
The first number on the table drags every later estimate toward it, even when the number is arbitrary; Tversky and Kahneman demonstrated it in 1974 with a rigged wheel of fortune. Every negotiation and every "was $89, now $49" tag runs on it.
3. Availability
Whatever comes to mind easily feels common. Plane crashes are memorable, so they feel likelier than the drive to the airport, which is the actually dangerous part. Vivid beats frequent in the mind's accounting.
4. Sunk cost
Money and years already spent argue for spending more, though they're gone either way. The degree, the renovation, the fourth season of the show. The clean question: knowing what I know now, would I start this today?
5. Negativity bias
Bad is stronger than good; one criticism outweighs five compliments at roughly that exchange rate. It kept ancestors alive and keeps you rereading the one curt email. Naming it helps more here than anywhere, because the distortion is emotional, not logical.
6. Hindsight bias
After the outcome, the past reorganizes itself into inevitability, and you always knew it. The cost is real: you stop learning from decisions because they all look obvious in the rearview. A decision journal (three lines at decision time) is the honest antidote; our journaling guide covers the format.
7. The planning fallacy
Your projects will take longer than you think, and knowing this does not fix it; Kahneman called it the bias he'd most like to cure in himself. The fix that works is outside-view estimation: ignore your plan, ask how long this kind of project takes people in general.
8. Dunning-Kruger
From the 1999 studies: the least skilled overestimate themselves most, because the skill needed to perform is the same skill needed to judge performance. Its popular use ("everyone dumber than me") is, pleasingly, an example of itself. The finding's real lesson points inward: confidence is not a credential, including yours.
Living with the eight
Debiasing has a mixed research record; checklists and outside views help, willpower doesn't. Slow, structured decisions where stakes are high, and forgiveness everywhere else. The full story of the two-system mind is in the books from our psychology starter guide, and 365 Days of Psychology walks the field one page a day, biases included, at a pace that gives each shortcut a day to be noticed in the wild.
Common questions
What's the most common cognitive bias?
Confirmation bias is the usual answer: it operates on every belief you hold, continuously, and feels like ordinary reading and remembering rather than bias.
Can you get rid of cognitive biases?
Not really; they're built into fast thinking. What works is structural: checklists, outside-view estimates, pre-committed criteria, and slowing down the handful of decisions that matter.
Is Dunning-Kruger real?
The 1999 effect replicates, with lively statistical debate about its size. The durable core: skill and self-assessment use the same competence, so the unskilled lack the tool to notice.
Where should I read more about biases?
Kahneman's Thinking, Fast and Slow is the definitive account by one of the field's founders. Our psychology books guide lists the friendlier on-ramps.