The Day Knowing Too Much Became a Problem

When experience stops helping

The scene is familiar and not particularly dramatic. A seasoned doctor, a chess player with thousands of games behind them, a highly skilled engineer, a senior executive with twenty years of experience. Competent people, with proven track records and long histories of good decisions. People who genuinely know what they are doing. And yet, when the environment changes, something starts to go wrong.

The rules shift. New tools appear. Signals change. Criteria for success are rewritten. What used to work almost automatically no longer does. At that point, something slightly uncomfortable happens: the people who adapt best are often not the experts, but the beginners. Novices ask more questions, hesitate more, and lack a single “right” way of doing things that needs defending.

For a long time, experience has been a powerful advantage. In stable environments, accumulated knowledge sharpens intuition, speeds up decisions and reduces mental effort. Doctors recognise patterns quickly, chess players “see” moves without calculation, and professionals act almost instinctively. Everything works smoothly, as long as the world stays more or less the same.

The problem begins when the ground starts to move. In medicine, when protocols change. In chess, when dominant styles evolve. In technology, when a new tool replaces ten older ones. In artificial intelligence, where what you learned two years ago can already feel outdated. In these contexts, experts often learn more slowly than beginners. Not because they are less capable, but because they already know.

Their mental models are deeply ingrained. Shortcuts that worked for years keep firing. Intuitions that were once reliable remain tuned to a world that no longer exists. Beginners, by contrast, have very little to unlearn. They are not defending a “correct” way of doing things, nor is their professional identity under threat. They try, fail, adjust and move on.

This phenomenon is known as the expertise reversal effect. It was described by Kalyuga, Ayres, Chandler and Sweller in 2003 as part of research on cognitive load. Their work shows that instructional methods that help novices learn can actually hinder experts when conditions change, precisely because experts already rely on highly optimised mental schemas built for a different environment.
👉 https://www.tandfonline.com/doi/abs/10.1207/S15326985EP3801_4

The lesson is not that expertise is useless. That would be a lazy conclusion. Experience remains valuable, but only under certain conditions. It accelerates learning and decision-making in stable worlds, and slows them down in fast-changing ones. That is why the most costly mistakes are often made not by those who know too little, but by those who are too confident in what they already know.

In environments that evolve quickly, such as data, artificial intelligence or complex decision-making, the most valuable skill is not accumulating more experience, but knowing when to stop trusting it blindly. Questioning assumptions, revisiting mental models and accepting that parts of your knowledge no longer apply is uncomfortable and rarely prestigious. But it works.

Sometimes, that is enough to keep learning while others quietly fall behind.