Omission Bias: A 6-Question Inaction Test Before You Decide

Which would you regret more: a mistake you made, or a mistake you allowed? Most of us answer quickly. The thing we did feels heavier. The thing we let happen feels like bad luck, or someone else's problem, or simply the way things went.

That instinct has a name: omission bias, the tendency to judge harm caused by an action as worse than the same harm caused by doing nothing.

In 1991 Mark Spranca, Elisa Minsk and Jonathan Baron put the instinct under a microscope. They wrote short stories in pairs. In each pair the person's goal, knowledge and the harm that followed were the same. Only one thing changed: in one version the person caused the harm by acting, and in the other they knowingly let it happen by staying silent. In one well-known scenario, a tennis player knows his opponent is allergic to an ingredient in the house dressing. In one version he recommends the dressing. In the other, the opponent orders it himself and the player says nothing. Many readers judged the player who recommended the dressing more harshly, even though he wanted the same outcome and got it.

Here is where it gets interesting. These were written scenarios, and not everyone showed the bias. The study does not prove that acting is always better than waiting, or that every choice to hold back is a mistake. It shows something narrower and more useful: when consequences are held equal, many of us still weigh them unequally, depending on whether our hands touched them.

Omission bias lives close to a few neighbours, so it helps to draw the lines. Status quo bias is about the pull of the current default, whatever it is. Omission bias is about responsibility: harm from inaction feels less like ours, even when there is no default at all. Loss aversion is the weight we give to losing something compared with gaining it. Omission bias can appear when gains and losses are perfectly balanced. And outcome bias judges a decision afterwards by how it turned out. Omission bias distorts the choice before anything has happened.

Why does this matter in practice? Because many real decisions are a choice between two kinds of risk: the risk of acting and the risk of not acting. If one kind feels heavier only because it would be visibly ours, we will lean toward waiting more often than the facts justify. The fix is not to always act. It is to make inaction carry its own name, its own consequences and its own signature.

Six questions for when one option is to act and the other is to wait:

1. What exactly happens if I do nothing? Write it as a sentence that starts with "I", such as "I leave the bug in place." Inaction that has a subject is harder to treat as neutral.
2. What harm could each option cause, in the same units? Money, time, people affected, chance of it happening. Put both branches side by side on the same scale and the same horizon.
3. If the harms were equal, which would I blame myself for more? Notice the answer. That gap is the bias, not information about the decision.
4. Who would see each harm, and would they know it was my choice? Visibility is not the same as responsibility. Separate how a result would look from what it would cost.
5. What would I advise a colleague in my place? Sometimes the answer comes more easily when it is not your name on the choice. Borrow that view for a minute.
6. What will I record before the outcome is known? Write down both branches, the estimates and your reasons in a decision journal, so the result cannot rewrite them later.

Suppose you lead a small software team. A bug occasionally corrupts customers' saved work. You have a fix ready, but any release carries some small risk of breaking something else. Doing nothing: "I leave the bug in place for another month." Same units: suppose the bug affects a few dozen customers a month, while the release, after testing, has a small chance of a short outage. Which would you blame yourself for more? Probably the outage, because you shipped it. Who sees what: the outage would be visible and traceable, while the corrupted files arrive one quiet support ticket at a time. Advice to a colleague: you would likely tell them to ship with a rollback plan. So you do, and you record both branches and your estimates in the journal. The numbers here are illustrations. The move is the point: give the quiet harm a voice before you choose.

But hold on. Sometimes waiting really is the better choice, when acting is irreversible, the evidence is thin, or doing something mainly serves to look busy. The six questions do not push you toward action. They make sure that if you wait, you wait on the merits. For a wider scan of related traps, the cognitive bias checklist is a good companion.

This is where ClearMind fits, quietly. It will not tell you whether to act or wait. It works as a mirror: when it guides you through Second-Order Thinking, it asks what happens next and what happens after that. You can answer for both branches, acting and waiting, and see what it reflects back about your patterns and blind spots. The choice, and the responsibility for it, stay with you.

Confucius said that to see what is right and not do it is a want of courage. Sometimes it is not courage we lack. It is noticing that standing still was a choice too.