# The small desk in your head > Why you can only juggle a few things at once, and how experts seem to hold so much more. Lesson 2 of 6 in [Why you remember and forget](https://learnfold.app/learn/memory/), a free course from Learnfold. Read this string once, then look away and say it back. F B I C I A U S A I B M Hard? Now try it as FBI, CIA, USA, IBM. They're the same twelve letters. This time you're holding four things instead of twelve, and four is about what fits. ### Working memory Working memory is where you hold things while you use them. It keeps a phone number alive between reading it and dialing it, and it holds the start of a long sentence while you wait for the verb. Think of it as a desk, with long-term memory as the filing cabinet. Everything on it is in reach, but the surface is small, and things slide off within seconds once your attention moves. In 1956 the psychologist George Miller wrote a famous paper about "the magical number seven, plus or minus two." People can repeat back about seven digits, so seven became the popular figure for how much we can hold. Later research told a messier story. When people repeat digits, they quietly rehearse them and group them into chunks. When experiments block those tricks, the limit comes out lower, around three to five chunks. The current best estimate is about four. Some researchers argue it isn't a fixed number of slots at all, but a pool of attention that thins out the more you spread it. ### Chunks change everything A chunk is any unit your long-term memory already knows. FBI is one chunk to you and three random letters to someone who has never heard of it. The limit is counted in chunks, which is why knowledge stretches it so far. Hide the letters and try to type them, then switch to familiar chunks. Practice affects later attempts, so use this to explore grouping rather than measure your capacity. The best demonstration comes from chess. Researchers showed players a board from a real game for a few seconds, then asked them to rebuild it from memory. Masters placed most of the pieces correctly. Beginners managed a handful. Then the researchers scattered pieces at random, in positions no real game would produce. The masters' advantage mostly vanished. Their desks weren't bigger. Where a beginner saw twenty separate pieces, a master saw a few familiar patterns, each one a single chunk pulled from years of play. In most fields, experts don't have more working memory. They have more stored chunks, so each spot on the desk holds more. ### Why this matters for learning When you're new to a subject, everything is a separate piece. A beginner reading about cells has to hold *mitochondria*, *ATP*, and *membrane* as strange words while also following the argument. The desk overflows and the explanation slides past. It feels like you're not smart enough. Usually you're out of space. A few ways to work with the limit: - Learn the key terms first. Once a word is a single chunk, it stops eating space. - Write things down and draw diagrams. Paper is an extension of the desk. - Study a worked example before you try a problem cold. For beginners, following a solved problem often teaches more than struggling with a blank one, because there's room left over to notice the pattern. - Do one thing at a time. Every switch between tasks clears part of the desk. The desk itself doesn't grow much with practice. What you can fit on it does, and that's what getting good at something feels like from the inside. ### Rabbit holes > Why did the chess masters lose their advantage on random boards? > Can working memory be trained, or do brain-training games only make you better at the games? > How do researchers measure working memory without letting people chunk or rehearse? Read online: https://learnfold.app/learn/memory/working-memory/