LOGBOOK

HELP

1 / 25
Other keys: showSpace: good1-4: rate0: skip5: flag

Question

What is a cognitive bias, and how does it differ from a logical fallacy?

Answer

A cognitive bias is a systematic, predictable error in how the mind perceives, remembers or judges; a logical fallacy is a flaw in the structure of an argument.

The two are easy to confuse because both lead to wrong conclusions, but they live in different places:

Cognitive bias Logical fallacy
Where it sits In your thinking/judgement In an argument's structure
Nature A built-in mental shortcut that misfires A reasoning step that doesn't follow
Example Believing a claim more because you've heard it often "Everyone believes it, so it's true" (ad populum)
Fix Notice it and consciously correct for it Point out the broken inference

A bias is something your brain does to you automatically — a side-effect of shortcuts (heuristics) that are usually efficient but predictably wrong in certain situations. A fallacy is something in the words of an argument you can lay out and inspect. The same mistaken belief can involve both: a bias makes the conclusion feel right, and a fallacy is the broken reasoning you'd use to defend it.

Tip: Biases are about the thinker; fallacies are about the argument. "Know thyself" targets biases — you can't audit an argument you're too biased to question.

Go deeper:

The Cognitive Bias Codex
The Cognitive Bias Codex
design: John Manoogian III categories and descriptions: Buster Benson implementation: TilmannR · CC BY-SA 4.0 · Wikimedia Commons
or press any other key

Question

What is confirmation bias, why do our minds do it, and what does it look like in practice?

Answer

The tendency to seek, notice and remember evidence that supports what you already believe, while overlooking or discounting evidence that doesn't.

How a prior belief filters incoming evidence

* A prior belief acts as a filter: confirming evidence is accepted easily, contradicting evidence scrutinised or ignored. *

The mechanism is that holding a belief is comfortable and revising one is costly. Once a belief is in place, the brain treats it as the default and grades incoming information against it: confirming evidence is accepted easily, contradicting evidence is scrutinised harshly or filtered out before it registers. We're effectively building a case for a verdict we've already reached, not testing a hypothesis.

Example: You decide a new colleague is unfriendly. Afterwards you vividly notice every curt reply, and the times they were warm or helpful slide past unremembered — so the evidence "piles up" in favour of a conclusion you formed on day one.

Tip: The antidote is to deliberately go looking for what would prove you wrong. If you only ever find agreement, you may simply not be looking for disagreement.

Go deeper:

If the true rule (T) encompasses the current hypothesis (H), then positive tests (examining an H to see if it is T) will not show that the hypothesis is false.
If the true rule (T) encompasses the current hypothesis (H), then positive tests (examining an H to see if it is T) will not show that the hypothesis is false.
MartinPoulter · Public domain · Wikimedia Commons
or press any other key