The Correction Scar

The correction scar is a metaphor this site uses in entries about illusions. It is not a scientific term. It names a systematic error that appears when a kind of processing that usually helps, such as allowing for shadows or comparing a region with its neighbors, meets a stimulus it does not fit.

Stated without the metaphor: under specific conditions, one processing step shifts people's judgments in a consistent direction, and experiments can measure the size of that shift.

What a scar can tell you

A scar shows where something happened, not how. In the same way, the error in an illusion suggests which processing step researchers might look for. It does not identify the mechanism.

Settling the mechanism takes experiments. Until they decide the question, competing explanations are set out side by side, and the metaphor is never offered as evidence for any one of them.

What it does not mean

The term does not imply that perception is otherwise free of error. A scar marks one place where a usually helpful step goes wrong; it says nothing about how often the same step goes right.

The term says nothing about purpose either. It does not claim that any processing step exists for a reason, or that the error is a price paid for some benefit.

Where it appears

The term is used only in entries about illusions.

In the Müller-Lyer entry, the step in question is the adjustment of perceived size for distance proposed in Gregory's corner account. That entry sets out four accounts of the illusion, and the corner account is only one of them.

In the checker-shadow entry, the steps in question are allowing for shadows and comparing a region with its neighbors, on this site's own lightness stimulus. That entry also sets out four accounts side by side, and none of them has been tested on its figure.