Fifteen cognitive biases and the moments they show up

Bias versus fallacy

The APA Dictionary defines a cognitive bias as a systematic error in thinking and a fallacy as an error in reasoning. The difference is where the error sits. A fallacy is in an argument and can be pointed at in a sentence. A bias is in the process that produced the judgement, often without any argument being made, and the person experiencing it usually feels that they are simply seeing things as they are. That is why biases are hard to catch in yourself and easy to see in others, which is itself one of the biases below.

Biases in what we notice

• Confirmation bias: seeking and weighing evidence that supports what you already believe. Trigger: any belief you would be embarrassed to drop.
• Availability: judging how common something is by how easily examples come to mind. Trigger: vivid news stories.
• Anchoring: letting the first number you hear pull every later estimate toward it. Trigger: an opening offer or a list price.
• Framing: reacting differently to the same fact stated as a gain or a loss. Trigger: "90% survive" versus "10% die".
• Salience: overweighting whatever is loudest or most recent. Trigger: the last thing someone said in a meeting.

Biases about ourselves

• Overconfidence: being surer than your track record justifies. Trigger: any forecast you have not written down.
• Hindsight: remembering an outcome as more predictable than it was. Trigger: "I knew it all along."
• Fundamental attribution error: explaining others' behaviour by character and your own by circumstances. Trigger: someone else's lateness.
• Self-serving attribution: taking credit for successes and blaming failures on the situation. Trigger: performance reviews.
• Illusion of explanatory depth: believing you understand how something works until asked to explain it step by step. Trigger: a policy you feel strongly about.

Biases about odds and losses

• Base-rate neglect: ignoring how common something is in general when judging a specific case. Trigger: a positive test for a rare condition.
• Gambler's fallacy: expecting independent events to even out. Trigger: a run of one colour.
• Sunk cost: continuing because of what has already been spent. Trigger: a project past its deadline.
• Loss aversion: weighing a loss more heavily than an equal gain. Trigger: selling anything you own.
• Survivorship: drawing lessons only from the cases that lasted. Trigger: advice from successful founders.

What actually helps

Knowing the list has a modest effect on its own; people who can name biases still commit them. The habits with better evidence change the situation rather than the mind. Write the forecast down before the outcome so hindsight cannot rewrite it. Look up the base rate before judging the case. Run a pre-mortem: assume the plan failed and explain why. Ask what would change your mind and go looking for that. Decide important things when rested, and not immediately after hearing a number.

Where this fits on the site

The fallacy test on this site trains noticing errors in arguments; biases are the errors that produce those arguments before anyone speaks. Neither the list nor the test measures rationality or intelligence. They give you names for patterns, which is the first and smallest step.