Fundamental attribution bias is the cognitive bias that leads us to attribute people’s actions, good or bad, to their character rather than to circumstances. A person of whom we have no opinion and for whom we have no ingroup/outgroup associations leading to an opinion, but whom we learn has done a bad thing, we generalize ((Remember: When we form our cognitive map of the world we generalize, distort, and delete.)) as a bad person; such a person who does a good thing, we generalize as a good person.
This is why you get at most one chance to make a first impression.

Once we have generalized a person as good or bad, we look at everything that he does in light of that generalization. Good people get the benefit of the doubt when they do bad things; bad people do not get the benefit of the doubt when they do good things.
This is the battleground in many criminal jury trials.

Something bad has happened. The State wants to pin it on the defendant, turn him into a bad person, and then get the jury to hold him culpable because he is a bad person. The defense wants to disrupt this chain, either preemptively (for example, with ingroup associations) making the defendant a good person, keeping the State from pinning it on the defendant, or fighting the uphill battle of showing the jury that the defendant is a good person despite the bad thing.

Specific example: self-defense killing. If defendant ended another human’s life (generally frowned upon) then he did so in self-defense. State wants to prove that he did it and take this fact to make him a bad person in the jury’s eyes, so that his self-defense case does not get the benefit of the doubt. The defense wants to show the jury from the beginning that the defendant is likable; or keep the State from proving that the defendant did a bad thing ((Pick a horse: “He didn’t do it,” or “It needed doing.”)); or, as a last resort, overcome fundamental attribution bias to convince the jury that, despite the bad thing, the defendant is not a bad person.
Good luck with that last one. Fundamental attribution bias, like all of our cognitive biases, is a powerful force with deep roots and adaptive evolutionary value. Like all of our cognitive biases, it’s not something that, in the normal course of our lives, we recognize having. ((What use is a bias that you recognize having?))

People ask me occasionally how I became a criminal-defense lawyer. I’ve never really had a good answer. There were no criminal-defense role models in my early life; I didn’t watch a lot of criminal-defense TV shows. “I just always tended to defend people,” was the best I could say. I remember pretty early on riding in a car with my mom driving, on I-270 in the DC area. Some other driver did something boneheaded, and Mom voiced her displeasure with him: “What a jerk!”

I pointed out that he was probably just having a bad day. Might be taking someone to the hospital. Might really need to go to the bathroom. ((Are there things that make people particularly good at their avocations that are not really annoying to their friends and family?))

When we are helping others overcome their biases, it helps if ours control us less than theirs control them. ((Requisite variety! A recurring theme here.)) And as criminal-defense lawyers, helping juries ((… and judges, and prosecutors, and witnesses, and other lawyers, and …)) overcome fundamental attribution bias, it helps us to have more of a handle on fundamental attribution bias in ourselves than most people have in themselves. If we are not tuned to see how circumstances could lead good people to do bad things, ((It works the other way too. There is good and evil in all of us.)) we aren’t going to do the bulk of our clients a whole lot of good.

So from now on that’s my origin story: “I’ve just always been less prone to fundamental attribution bias than most people.” ((I’m sure I have some other blind spot that more than compensates, making my world a worse place rather than better.))

You, at least, will understand.