Cognitive biases
Self-serving bias: why success feels like ours and failure feels like chance
Every evening Julia and her brother play a phone game called “Quick Eye.” Blue circles and orange squares flash on the screen for a moment, and you have to say which there were more of. On Monday Julia gets 4 out of 5. “I just have a sharp eye,” she says. On Tuesday she gets 1 out of 5. “The shapes were kind of tricky today, and my phone was lagging.” Same game, same Julia. But completely different explanations.
This is self-serving bias: we tend to explain our successes by our abilities and effort, and our failures by circumstances, chance, or other people’s mistakes. In part, this makes it easier to keep a good opinion of ourselves. But there is a price: in a failure it is harder to see our own share and to learn something from it.
Below, we look at what studies have shown, why this habit has a useful side too, and how to look at your failures honestly but without beating yourself up. In the middle of the article you will play “Quick Eye” yourself.

Contents
In short: how self-serving bias works
What self-serving bias is
Psychologists noticed long ago that when people explain their own successes and failures, they do not judge them the same way. “Self-serving” here does not mean “greedy.” It means that the explanation serves us: it helps us feel better.
At first the effect was doubted. In 1975 the psychologists Dale Miller and Michael Ross went through the studies available at the time. The paper was titled “Fact or fiction?” Their conclusion was cautious. The data partly supported that people credit themselves with success. But that people blame failure on circumstances had almost no support. And even the first half, the authors suggested, could be explained without any protection of self-esteem: people simply expect their efforts to lead to success. When that happens, the cause seems obvious: it’s me.
Then the effect was found many times. Four years later Miron Zuckerman went through the studies again and saw both halves of the effect in most, though not all, kinds of experiments. In his view, it is still about the wish to protect and raise self-esteem. A large summary from 2004 pooled 266 studies. People explained good outcomes by themselves noticeably more often than bad ones, and this tilt turned up in almost all groups of participants. It was strongest in children and in older people. Among participants from Asian countries the tilt was noticeably smaller than among participants from the US and other Western countries.
When the effect is stronger. According to a summary by Keith Campbell and Constantine Sedikides, the tilt grows when a failure stings our pride more. The more it hurts our opinion of ourselves, the stronger the tilt.
Try it yourself: “Quick Eye”
Now you are in Julia’s place. The game takes a couple of minutes. Here is what will happen.
Step 1. The first round. Five times, blue circles and orange squares will appear for a moment. Each time you answer which there were more of. Then you will see your score.
Step 2. You move a slider and decide what your score depended on: you or the circumstances.
Steps 3 and 4. The second round and the same question again.
Step 5. The results: we show both of your answers side by side.
Why do the shapes appear so fast? That way they can’t be counted, and you have to rely on your eye, like Julia. What the computer shows and in what order differs from player to player. Why, we will tell you at the end.
What scientists usually find. People more often explain successes by themselves and failures by circumstances. If after a good score your slider went further toward “it’s me” than after a bad one, that looks like the same tilt. After all, the rounds were the same, and the score did not depend on your answers.
What the game can’t show. One person and two rounds is a hint, not a test. You may not have believed the score, or you may have simply gotten tired by the second round. And someone who read the article before the game already expects a trick. If your answers almost matched, that is an honest result: the rounds really were the same.
When it is not a mistake. Sometimes a failure really does depend on circumstances: the phone really was lagging, the examiner really was strict. Looking for outside causes is normal. It becomes a mistake when all the causes of failures end up outside and all the causes of successes end up inside.
Why success feels like ours
Imagine you have a lawyer who is always on your side. When things worked out, he says: “That’s your achievement.” When they didn’t: “It’s not your fault, circumstances just fell that way.” He is pleasant to listen to. But if you want to learn something, you also need an honest coach. This is an image, not how the mind is built.
Scientists argued for a long time about where the “lawyer” comes from. Some explained everything by expectations: we expect success, and when it comes, we link it to ourselves. An unexpected failure is easier to explain by something outside. Others saw in the effect a wish to protect self-esteem. In a 2008 review, James Sheppard and colleagues concluded that there is no need to argue “either-or”: expectations and the wish to think well of ourselves usually work together.
There is also a careful question: might the “lawyer” have a useful side? In the 2004 summary the tilt was much weaker in groups of participants with depression or anxiety. Why, these data can’t say: the link could run in either direction. So there is no answer to that question here, and we draw no conclusions about people. Your result in the game says nothing about it: two rounds are far too few to judge anyone’s mood.
Where it gets in the way
School. “An A, because I’m smart; an F, because the teacher is picky.” If every failure is explained this way, it is unclear what to change in your preparation. The cause is always outside, and you can’t influence circumstances.
Teamwork. The project succeeded: everyone remembers their own contribution. It failed: everyone remembers who let the team down. In experiments by Sedikides and colleagues, people did tasks in pairs, and the result was shared. Partners who barely knew each other credited a shared success to themselves and a failure to the partner. Close people did not do this: they did not take more credit than their partner, and did not blame the partner more than themselves.
Arguments. After an argument it is easy to remember what the other person said and forget how you started it. This is an everyday observation. The experiments with partners hint at what may help: warmth toward the other person softens the habit of blaming them rather than yourself.
The flip side. It also happens the other way around: a person writes off every success as luck and every failure as their own fault. “I got lucky with the questions,” “I messed everything up again.” This is also a tilt, just in the other direction, and it can be hard to live with. Here we rely on an everyday observation, not on experiments. If you recognize yourself in this and it has been going on for a long time, it is worth talking to a specialist. Not because something is wrong with you, but because you deserve a fairer view of yourself.
How to look at successes and failures more honestly
Self-serving bias is one of the biases on the cognitive bias map, in the section on how we judge people and ourselves. Next to it is the Dunning-Kruger effect: how we assess our own knowledge.
Frequently asked questions
Is this the same as the fundamental attribution error?
No. The fundamental attribution error is about how we explain other people’s actions: we see character and miss circumstances. Self-serving bias is about our own successes and failures. The popular rule “we explain other people’s behavior by character and our own by circumstances” has held up poorly under testing; that is a separate story.
How is this different from hindsight bias?
Hindsight bias is about how, after an event, the outcome seems predictable: “I knew it.” Self-serving bias is about whom we give the credit or the blame.
Is it the same in every country and at every age?
According to the 2004 summary of 266 studies, the tilt toward explaining good outcomes by ourselves turned up in almost all groups of participants. It was strongest in children and older people, and noticeably smaller among participants from Asian countries than among those from the US and other Western countries.