Memory Palaces and the 64 Squares

In the 1970s, a pair of researchers showed chess positions to masters and novices for five seconds, then asked them to reconstruct what they had seen. The masters were dramatically better — until the pieces were placed at random, at which point the masters were no better than anyone else.

That result is the single most useful thing anyone has ever discovered about chess memory. Masters do not have superior memories. They have superior chunks.

Chunks, not squares

A novice looking at a position sees perhaps twenty-five separate facts: a piece here, a piece there. A master sees five or six: a castled king with an intact pawn shield, a standard isolated queen's pawn structure, a knight outpost, a rook doubled on the open file. Each chunk is one item, and each item carries a great deal of information because it comes from thousands of previous games.

This is why random positions destroy the advantage. Random arrangements contain no chunks, so there is nothing to compress.

Training the compression

Board Memoriser puts a position in front of you for a few seconds and asks you to rebuild it. The instinct, at the beginning, is to memorise square names — pawn on c4, knight on f3 — like a shopping list. It works for about four pieces and then falls apart.

The faster route is to force yourself to describe the position in relationships. These three pawns form a chain. This bishop and this queen are on the same diagonal. The king is uncastled and the centre is open. You will recall more pieces this way while consciously holding less.

When you make errors, look at which pieces you lost. Almost always they are the ones that belonged to no chunk — a lone rook on a random file, a pawn doing nothing. That is not a memory failure. It is a position with an incoherent piece in it, and noticing that is itself a chess skill.

What it gives back

Better memorisation is not the point. The point is that chunking is the same mechanism as positional understanding. Every time you compress a position into meaningful units, you are practising the thing that lets a strong player glance at a board and say 'White is better, the knight has no squares'.

Eight pieces, eight seconds, a few times a week. It is a strange exercise that pays out somewhere else entirely.