Original title: Perceptibility of Error Outcomes Shapes the Direction of Post-Error Adjustment: Evidence From 6.7 Million Moves in Online Chess
Yunjeong Jeong, Younggyu Kim · September 2026
Do people slow down after making a mistake only if they notice it? In typical lab tests, making an error and realizing it happen together, so scientists cannot tell them apart. Researchers solved this by studying chess. Computer engines can spot blunders that do not change the board, while losing a piece is immediately obvious. They studied how players reacted right after different types of mistakes. They found that whether an error was visibly obvious completely flipped how players reacted. When players made a mistake but lost no pieces, they slowed down on their next move. But when players visibly lost material, they actually moved faster. Ultimately, post-error reactions split into two patterns based on whether the outcome was visible.
Imagine you are hiking with a friend on a winding trail. At a fork, your friend confidently picks the left path, but you suspect it's a dead end. If the path ends in clear view ahead, you both quickly turn around. But if the path loops back hidden behind trees, you might hesitate and walk slower, unsure whether to change course. In this paper, the hiker's hesitation maps to the player's slower move after a hidden blunder, while the quick turnaround matches the faster move after losing a visible piece.
This matters because it challenges a core assumption in psychology: that people slow down after mistakes to be more careful. The study shows that only visible mistakes—like losing a piece—trigger faster moves, while invisible ones lead to slower moves. That changes how we think about learning from errors in anything from driving to medical diagnosis.
The abstract does not address whether these effects generalize to non-chess tasks or to other forms of error feedback. It also does not establish causality: the correlation between error visibility and adjustment direction could be driven by unmeasured third variables, such as individual differences in attention or skill. The reliability near zero for blunders suggests that imperceptible errors may not consistently elicit any systematic adjustment at the individual level.
Ask anything about this paper. Answers come from its abstract, and say so when the abstract doesn't cover your question.
Perceptibility of Error Outcomes Shapes the Direction of Post-Error Adjustment: Evidence From 6.7 Million Moves in Online Chess — Yunjeong Jeong, Younggyu Kim — PsyArXiv — doi:10.31234/osf.io/mvab2_v2
Abstract licence: CC BY
Explanation generated by Synthesis from the abstract — it is not the original paper. This is a preprint, so it hasn't necessarily been peer reviewed.