The map in front of you and the map in your head.

Picture Africa and Greenland. Write down their area ratio before checking the numbers. This separates the image you see from what you know about Earth. Mercator’s increasing area scale is a mathematical property; a cognitive map is an internal spatial representation. Studying the latter requires tasks involving people.

Section references[1][3]

The little map lab

How much Africa is in your mental map?

Estimate how many times Greenland’s total area fits into Africa’s. Move the slider, then reveal the ratio.

AfricaGreenland
Square areas in proportion, not country outlines. Side lengths change with the square root of the area ratio.

What researchers tested.

Battersby and Montello’s two 2009 studies used numerical and graphical area estimates. They found little or no Mercator-like influence. Lapon, Ooms and De Maeyer’s worldwide online study in 2020 likewise found no mental copy of a single projection, although estimation accuracy differed with representation.

Section references[1][2]

Three questions that need different evidence.

Is the drawn area distorted? Do people estimate differently? Does that change political decisions? Geometry answers the first; perception and memory tasks address the second. The third needs evidence about decisions. An area-estimation study does not directly establish a geopolitical consequence.

Ask what was measured, who participated and which alternative explanations were examined before accepting a broad claim.

Section references[1][2][3]

Why the task matters.

An isolated outline, a labelled world map and a drawing from memory are different tasks. Prior knowledge and the response method can matter. An open online sample also involves self-selection: large does not automatically mean representative.

These limits neither establish a strong universal effect nor prove maps irrelevant. Association alone does not identify long-term exposure as the cause.

Section references[1][2]

A useful habit for reading maps.

Record a prediction, compare it with an equal-area map and numerical data, then separate what changed on screen from what you learned. Our experiment is an educational exercise. It collects no research data and does not measure intelligence or political attitudes.