Chunking, spatial anchoring and a fixed reveal order — three techniques that reliably raise your Memory Match percentile.
Memory Match feels like a raw memory test, and that is why people assume they cannot improve at it. They are wrong. Working memory has a fixed capacity, but chunking lets you fit more into it.
These three techniques are the ones with the largest measured effect on our own leaderboard.
Chunk the board into quadrants
Do not try to remember sixteen independent cards. Remember four groups of four, and give each quadrant a rough label: "the top-left has two animals and two shapes".
Chunking works because recall of a group is far more reliable than recall of its individual members, and a partially remembered chunk still narrows your search.
Anchor on position, not on imagery
Players who try to memorise pictures do worse than players who memorise locations. Say "the sun is third row, second column" rather than holding the image in your head.
Positional memory uses a different system from visual memory and it is much better at surviving a few seconds of interference.
Use a fixed reveal order on the first pass
Reveal cards in a consistent sweep — left to right, top to bottom — rather than clicking whatever catches your eye. A fixed order means position carries information for free.
On the second pass you can deviate, because by then you are working from recall rather than encoding.
FAQ
Does playing more actually improve memory generally?
Practising a memory game makes you better at that game. Transfer to unrelated tasks is modest — see our piece on dual n-back and working memory for the honest version.