That Error in Your Head Is Not a Character Flaw

Line seven of the standard risk-assessment form reads: "Rate the likelihood of this outcome on a scale of one to ten." There is no column for when you last saw this outcome happen to someone you know, no column for how recently you read about it, no column for whether it frightened you. The form asks for a probability. It collects a feeling wearing a number.

A bias is a shortcut that served well in one environment and misfires in another — the shortcut is not the problem, the environment is. Your brain did not evolve to assess base rates in spreadsheets. It evolved to weight recent, vivid, personally witnessed events heavily, because in a world without records those events were the most reliable data available. That old office still runs its code. When the environment changes and the code does not, you get a systematic, predictable error. Predictable means workable.

Daniel Kahneman and Amos Tversky mapped this precisely in the 1970s: the availability heuristic causes people to overestimate the frequency of events that are easy to recall, not events that are actually common. Plane crashes feel more likely after news coverage. Rare diseases feel epidemic after a friend's diagnosis. The insight that followed was structural: because the error is consistent and directional, it can be anticipated. Checklists in aviation, second-opinion protocols in medicine — these are not willpower. They are environments redesigned around a known misfire.

You have a hard conversation coming up, probably before the end of the day. You are going in with a read on how the other person will respond, and that read is built from the last time something like this went badly. That memory is loud and recent, which means your brain is pricing it as more likely than the base rate warrants. Before you walk in, write one sentence: what outcome do you actually expect, and what is that expectation based on? Not as therapy. As calibration. The act of writing it forces you to distinguish between what you remember and what you know, and that gap is where better decisions live.

The form in line seven is not broken because humans are broken. It is broken because it assumed the person filling it out had already done this work. Most forms do. Most of us have not. That is the only thing worth fixing.