Cognitive Bias vs. Logical Fallacy: What's the Difference?
One corrupts how you think; the other corrupts how you argue, and confusing the two is its own kind of mistake.
People use cognitive bias and logical fallacy as if they were the same accusation hurled from two angles. They're not. They live in different parts of the machine.
Here's the clean line: a cognitive bias is a flaw in how your mind forms a judgment, a systematic glitch in perception, memory, or estimation that happens before you've said a word. A logical fallacy is a flaw in how an argument moves from premises to conclusion, a broken step in the reasoning you put on the table.
One is a bug in the camera. The other is a bug in the edit.
The bias: a flaw in how you think
Biases are upstream. They warp the raw input before you're even aware you're judging anything. They're not about being dumb; they're about being human-shaped, running on hardware optimized for the savanna and not for spreadsheets.
Four worth knowing by name:
- Confirmation Bias, you hunt for evidence that flatters what you already believe and quietly file the rest under noise. You're not lying. You're just steering. - Availability Heuristic, you judge how likely something is by how easily an example springs to mind. One shark documentary and suddenly the ocean is out to get you, while the staircase that statistically wants you dead gets a pass. - Anchoring Bias, the first number you see hijacks every number after it. A jacket marked down from $400 feels like a steal at $180, even if $180 is exactly what it's worth. - Dunning-Kruger Effect, the less you know about a domain, the less equipped you are to notice how little you know. Incompetence comes with its own built-in confidence subsidy.
Notice that none of these are arguments. They're distortions in the lens. You can hold a perfectly biased belief and never once make a logical case for it.
The fallacy: a flaw in how you argue
Fallacies are downstream. They show up in the structure of a claim, the move from "because this" to "therefore that." A fallacy is a reasoning step that looks valid and isn't.
Four to recognize on sight:
- Ad hominem, attacking the person instead of the point. "You can't trust her budget analysis, she's been divorced twice." Cool. The spreadsheet doesn't know that. - Straw man, swapping someone's actual argument for a flimsier version, then triumphantly knocking that down. They said "regulate it," you heard "ban it forever," and off you went. - False dilemma, presenting two options as if they're the only two. "Either we cut the program or we go bankrupt." There are usually doors three through nine. - Slippery slope, claiming one small step inevitably triggers a catastrophic chain, with no evidence the chain exists. Let kids have phones in class and clearly civilization ends by Thursday.
There are more, appeal to authority (it's true because someone important said so), circular reasoning (it's true because it's true). The common thread: the logic is broken, regardless of who's reasoning or what mood they're in.
The cleanest way to tell them apart
Strip it to one question: is the error in the head, or on the page?
- A bias is private and silent. It happens inside you, before language. You can be biased while saying nothing at all. - A fallacy is public and structural. It happens in the argument itself. You could feed it to a logic checker and watch it fail, no human required.
Put differently: a bias is why you reached a bad conclusion. A fallacy is how you defended it. One is the crime; the other is the alibi.
Where they blur
Here's the honest part: they feed each other constantly.
Confirmation Bias quietly generates fallacies. Once your mind has picked a side, it'll reach for a Straw Man to dismiss the opposition and a False Dilemma to corner you into the answer you already wanted. The bias chose the destination; the fallacy paved the road there.
Sunk Cost Fallacy is the great shapeshifter, and a useful warning about labels. It's named a fallacy, and as an argument it is one: "we've spent two years on this, so we have to finish." But the force behind it is a bias, loss aversion, the gut refusal to write off what's gone. The reasoning error and the perceptual error are wearing the same coat.
Even Survivorship Bias sneaks into arguments dressed as evidence. You point to three college dropouts who became billionaires and conclude that dropping out works, never counting the millions who tried it and vanished. The bias hid the data; the argument just walked through the hole it left.
The overlap isn't a failure of the categories. It's a reminder that flawed perception and flawed reasoning are partners, not strangers.
Why this matters for decisions
Because the fix is different depending on which one you've got.
You don't out-argue a bias, it never made an argument. You counter it by changing your inputs: seek the disconfirming evidence, check the base rate, distrust the first number, get a second pair of eyes that doesn't share your blind spot.
A fallacy, you fix by changing the structure: lay the premises out, check whether the conclusion actually follows, and refuse to be cornered by a fake either/or.
Mistake one for the other and you treat the wrong patient. You'll bring airtight logic to a perception problem, or fact-check a feeling that was never about the facts. Most bad decisions aren't a failure of intelligence. They're a failure to notice which layer broke.
---
You now know the difference in theory. The harder question is which one is yours, the flaw your mind reaches for first, the blind spot with your name on it.
There's a way to find out. Take the test and meet your signature blind spot.