Moonwalking with Einstein Summary & Review: Train a Memory You Can Use

Moonwalking with Einstein shows how an ordinary memory becomes usable — palaces, the baker/baker paradox, the OK plateau, and a 7-day plan.

★★★★½ 4.6/5 — a journalist trains an ordinary memory for a year and shows the skill was hiding in pictures, places, and harder practice.

Best for: Students, speakers, and knowledge workers who keep losing names, lists, and the structure of what they just read

Reading time: ~6.5 hours (320 pages) · this guide ~16 min

Difficulty to apply: Medium — the palace is simple; ten minutes of encoding a day is the part most people skip

Moonwalking with Einstein in one minute

You do not have a bad memory. You have an untrained one. Joshua Foer covers the 2005 USA Memory Championship as a science journalist, meets people who memorize decks of cards for fun, and takes a dare: train for a year. With British champion Ed Cooke as coach, he learns the memory palace and the habit of turning facts into images strange enough to find again. He returns in 2006 and wins. The stunt is the plot. The useful book is about attention: why a name called Baker slips while a baker sticks, why practice dies on the OK plateau, and why outsourcing every fact to a phone also outsources the looking that makes a fact yours.

Key takeaways

  1. Memory is a skill, not a trait: Foer starts average and wins a national title. The hardware is ordinary. The encoding is not.
  2. We outsourced remembering: Writing, then search, then the phone. External memory is powerful. It also lets you skip the work of noticing.
  3. The palace is a walk you already know: A house, a commute, a school hall. Place one image per stop. Recall is replay, not rummaging.
  4. Weird beats tidy: A polite mental note fades. A sumo wrestler in the kettle does not.
  5. Baker is not baker: A name is a thin label. A job comes with flour and heat. Give facts meaning or they leave.
  6. The OK plateau is where most practice dies: Once a skill feels easy, hours stop buying improvement unless you make practice hard again.
  7. Deliberate practice needs a target and a clock: Specific drill, fast feedback, slightly past comfortable. Showing up is not getting better.
  8. Chunking multiplies capacity: Experts hold richer packets — a hand of cards, a line of verse, a face plus a story.
  9. Championship speed is optional: You need the palace for names and talks. You do not need a shuffled deck.
  10. A trained memory is a way of paying attention: Encoding forces you to look. That is the transfer that survives the last page.
Chart of the OK plateau: skill rises, flattens on autopilot, then climbs again with deliberate practice
Source: Moonwalking with Einstein by Joshua Foer · Chart © thegrowthreads.com

Moonwalking with Einstein by Joshua Foer book cover
Cover © Penguin Books. Used for review and identification.

What is Moonwalking with Einstein about?

Moonwalking with Einstein is a reported memoir about how human memory actually works and how to train it. Joshua Foer spends a year learning ancient mnemonic techniques, explains the science behind images and places, and uses his own run at the USA Memory Championship to show that remembering more is a practice you can install, not a gift you either got or missed.

About the author

Joshua Foer was born in Washington, D.C. in 1982 and trained as a journalist, writing for National Geographic, Esquire, Slate, and the New York Times. In 2005 he was assigned to cover the USA Memory Championship as a curiosity. Instead he met mental athletes who insisted their results came from technique, not talent, and he took a year of coaching from Ed Cooke. He won the 2006 title, then turned the experiment into this 2011 Penguin book. Foer later co-founded Atlas Obscura and the Sukkah City design competition. The through-line is attention: how people notice and retrieve what the rest of us scroll past. Moonwalking with Einstein is his first book and still the cleanest popular account of trained memory as a learnable craft. Explore all Joshua Foer book summaries →

Key concepts at a glance

Concept What it means Use it when
Memory palace A familiar route used as a filing cabinet for images You need a list, a talk, or a poem in order
Method of loci The classical name for placing facts along a path You keep losing sequence, not just items
Elaborative encoding Attach meaning, senses, and story so a fact has hooks A name or date will not stay overnight
Baker/baker paradox A job sticks; a same-sounding name does not You meet people and blank on the label
OK plateau The flat stretch after you get good enough Hours of practice no longer move the skill
Deliberate practice Hard, specific, feedback-rich drill past comfort Autopilot feels productive and nothing improves
Chunking Pack bits into meaningful groups A deck, a verse, or a formula feels like too much
External memory Notes, search, and phones that hold what you do not You need volume and search, not instant recall

Part 1: We outsourced memory — and forgot the skill

Foer arrives at the championship as a tourist. The room looks like a spelling bee for adults who like playing cards. Then the scores land and the story flips. These are not savants reciting pi because a film said they should. They are people who built a filing system and practiced it. Cooke’s version is blunt: anyone who cares enough can get on the board. Foer takes that as a reporting assignment and, against his own caution, as a training plan.

The history chapter changes how you see your phone. Before cheap writing, memory was a civic skill. Orators walked imagined buildings because a speech had to live in the body. Books made forgetting cheaper. Search made it cheaper again. Foer is not nostalgic for a world without paper. He is precise about the trade. External memory is a gift for volume. It is a loss if you never encode the few things that should be available when the battery is in the other room.

Early training is clumsy. Lists vanish. Palaces collapse because the route is vague. Cooke sends him back to a house he can walk in the dark. The first lesson is not a trick. Remembering starts before the test: you decide a fact is worth a picture. Most of us hear a name, nod, and hope. Hope is not a memory system.

TGR Note: To get more from hard nonfiction, start with How to Read a Book and Make It Stick. They train inspection and retrieval. Foer trains the other half: making a fact pictorial so retrieval has somewhere to land.

Three-step Memory Palace diagram: pick a place, drop weird images, walk the route
Source: Moonwalking with Einstein by Joshua Foer · Diagram © thegrowthreads.com

Part 2: Build a palace and make it weird

The method of loci is older than the book and simpler than the name. You pick a place your body already knows. You fix a path. At each stop you drop an image that stands for the thing you want later. To recall, you walk. Foer’s championship images are lurid because bland pictures evaporate. The swimsuit and wrestler on the paperback are the technique printed on the jacket.

Elaborative encoding is the science under the joke. Tell someone the person is a baker and the brain gets an oven, flour, heat. Tell them the last name is Baker and the brain gets a label. Same sound. Different fate. Every name you lose at a conference is a Baker you never turned into a baker. Foer’s fix is not “try harder.” Give the syllable a scene and stick it to a face or a doorway.

Chunking shows up when lists get long. A mental athlete does not hold fifty loose digits. They hold packets: one image for a pair of cards, one scene for a three-digit run. You already chunk phone numbers. Do it on purpose for a talk, a client list, the headings of a paper. Competitive PAO systems are optional. One rich picture still beats five thin words.

TGR Note: Limitless covers overlapping mnemonics with more “brain optimization” language. Building a Second Brain is the opposite tool: store volume outside your head. Use a second brain for files you will search. Use a palace for the five names and five points you refuse to read off a slide.

Baker/baker paradox diagram: a name is a thin label, a job is a picture the brain can keep
Source: Moonwalking with Einstein by Joshua Foer · Diagram © thegrowthreads.com

Part 3: The OK plateau is where most practice dies

The most portable idea in the book is not the palace. It is the plateau. Skill has three stages: cognitive (you think through every step), associative (errors drop, speed rises), autonomous (you can do it without watching yourself). Autonomy feels like mastery. It is often just comfort. Typists and language learners know the shape: a steep early curve, then years of “practice” that do not move the score.

Anders Ericsson’s deliberate practice is the way through. Pick the sliver that still fails. Design a drill that makes the failure loud. Get feedback before the next attempt. Stay a little past comfortable. Foer does this with cards and poems: timed recalls, weak loci rebuilt. The championship is a scoreboard. The method works on a Tuesday for a talk you keep muddling.

The book stays honest about talent. Some people start with a faster visual flicker. Foer does not pretend a year of palaces turns every reader into a champion. He does argue that the gap you blame on talent is often a gap in encoding and in the quality of practice. If your hours are unmeasured and your errors are invisible, you are living on the plateau and calling it experience.

TGR Note: Peak is the full Ericsson brief — read it when you want the research spine without the championship plot. Ultralearning and So Good They Can’t Ignore You apply the same hardness to career skill. Foer is the friendlier on-ramp: one palace, one plateau, one year of proof that ordinary practice can be redesigned.

Three-stage OK plateau diagram: cognitive, associative, then autopilot where improvement dies
Source: Moonwalking with Einstein by Joshua Foer · Diagram © thegrowthreads.com

Part 4: What a trained memory is actually for

The last movement refuses the party-trick ending. Foer wins. He also asks what the win is worth. A mind that can hold a deck is impressive at a banquet and useless if it never looks at the people at the table. Encoding is a form of attention. When you build an image, you have to see the thing. When you walk a palace, you have to visit it again. That loop is closer to how experts live inside a field than to how contestants live inside a stopwatch.

He visits researchers and a few extreme cases — people who cannot forget a date, people whose memory is damaged — so the book does not collapse into a pamphlet. The contrast is the point. You do not want every day stored forever. You want the freedom to choose what stays. A trained memory is selective on purpose. It is the opposite of a feed that keeps everything and teaches you nothing.

Foer writes that monotony collapses time and novelty unfolds it. If your workdays are identical, your memory of them will be too. A palace injects novelty on demand. So does a name you finally attach to a scene, or a chapter you can close and still teach. You leave able to install a small system: one route, ten minutes, a slightly harder drill when the route gets easy. The medal is optional.

TGR Note: If the problem is a day that never stays on one object, Deep Work is the calendar version of a palace. Atomic Habits is how the ten-minute encoding block survives week two.

Who is Moonwalking with Einstein best for — and who should read something else first?

Best for students who reread and still blank, speakers who do not want to live on slides, managers who lose names, and anyone curious how a “bad memory” behaves once you give it a route. You do not need to care about card decks. You need a short list of facts that should be available without a screen.

Read something else first if you want the lab version of learning — start with Make It Stick. For storing files and quotes, use Building a Second Brain. To inspect a hard book before you memorize anything, start with How to Read a Book. If you cannot protect ten minutes, fix the queue with Getting Things Done or The Phoenix Project, then come back.

Questions to reflect on

  • Which five facts this month would be worth a palace stop — and which ones should stay in a notebook?
  • Which names do you lose, and what picture would turn each Baker into a baker?
  • Where are you on the OK plateau in a skill you still call “practice”?
  • What would a 10-minute encoding block look like if it had a clock and a score?
  • What do you currently outsource to a phone that you wish you could retrieve in a hallway?

🔥 Ready to train a memory you can actually use?

Foer’s year is the story. The palace and the plateau are the parts you can run this week — then use the 7-day plan below.

Get it on Amazon
Bookshop.org
Audible

How to apply Moonwalking with Einstein (7-day plan)

  1. Day 1 — Pick a palace Walk your home and mark ten fixed stops: front door, coat hook, kettle, desk. Write the route once. Do not invent a castle. Use a place your feet already know.
  2. Day 2 — Encode five items Take today’s to-do list or a grocery list. Place one vivid, slightly ridiculous image at each of the first five loci. Walk the route twice from memory before you shop or start work.
  3. Day 3 — Baker a name Pick one person whose name you keep losing. Turn the sound into a picture and put it on their face or at a palace stop. Use it the next time you see them.
  4. Day 4 — Add a second room Extend the palace by ten loci, or add a commute. Practice the old five items, then place five new ones. If a stop is fuzzy, make the image louder, not the route longer.
  5. Day 5 — Encode a short talk or poem Choose a five-point outline or a stanza. One image per point, in order. Recite while walking the palace. Time the first clean recall.
  6. Day 6 — Find the weak locus Recall yesterday’s set. Circle the stop that failed. Rebuild that image so it is uglier, funnier, or more physical. Timed recall is the feedback loop.
  7. Day 7 — Install ten minutes Put a daily 10-minute encoding block on the calendar. Same palace, new list, or a name. The habit is the point. Championship speed is optional.

Frequently asked questions

What is Moonwalking with Einstein about in one paragraph?

Moonwalking with Einstein is Joshua Foer’s first-person account of spending a year training his ordinary memory with techniques that date to ancient orators. After covering the 2005 USA Memory Championship as a journalist, he apprenticed with British memory champion Ed Cooke, learned the memory palace and elaborative encoding, and returned to win the 2006 title. The book mixes that plot with the science of why names slip, why practice stalls on the OK plateau, and why a trained memory is less about party tricks than about paying attention on purpose.

What is a memory palace in Moonwalking with Einstein?

A memory palace is a place you already know — a house, a commute, a school hallway — used as a filing cabinet for images. You walk a fixed route and drop a vivid, often ridiculous picture at each stop. To recall the list, you walk the route again. Foer uses this method of loci to store cards, poems, and names. The palace is not magic. It works because spatial memory is older and stickier than raw verbal memory, and because a weird image is harder to lose than a string of letters.

What is the baker/baker paradox?

The baker/baker paradox is a simple experiment with a large lesson. People remember that someone is a baker more reliably than they remember that someone’s last name is Baker. The job comes with flour, heat, and a white hat — hooks the brain can keep. The name is a thin label. Foer uses the paradox to explain elaborative encoding: if you want a fact to stay, give it meaning, sensory texture, and a picture. A name you keep forgetting is usually a Baker you never turned into a baker.

What is the OK plateau?

The OK plateau is the flat stretch after you get good enough. Early practice is cognitive: you think through every step and errors are obvious. Then you become associative: faster, cleaner, less effort. Then you go autonomous — autopilot — and improvement stops even if you keep putting in hours. Foer, following Anders Ericsson, argues that experts stay in the uncomfortable middle: they design practice that is specific, feedback-rich, and slightly too hard. Showing up is not the same as getting better once the task feels easy.

Do you need a special brain to use these techniques?

Foer’s own experiment is the answer he offers: he started as a reporter with a normal memory, trained for a year, and won a national championship. The mental athletes he meets are not a separate species. They use the same spatial and visual hardware everyone already has. Savants and people with extraordinary autobiographical memory appear in the book as contrasts, not as the model. The practical claim is modest and useful: you can remember more of what you choose to encode. You cannot download a new hippocampus.

How long does it take to build a usable memory palace?

A first palace can be useful the same afternoon. Pick ten loci in your home, encode a five-item list, and walk it twice. Most readers feel the difference on a grocery run or a talk outline within a week if they practice ten minutes a day. Championship speed — a shuffled deck, a long poem — takes months of deliberate, timed drills. Foer’s year is the high end. You do not need his year to stop losing names at a conference. You need one route and the habit of making the image strange enough to find again.

Should I still use notes if I train my memory?

Yes. Foer is not arguing that you should memorize your calendar or throw away a notebook. External memory — paper, a phone, a second brain — is excellent for volume and search. A trained memory is for the few facts you want available without a screen: names, a talk, a poem, the structure of an argument. Use both. The book’s warning is narrower: if every fact lives only in a device, you also outsource the attention that encoding would have forced. Notes hold. Palaces make you look.

Related summaries

How we analyze books: every book on The Growth Reads is read cover to cover, summarized in thousands of words of original analysis, and rated against our five-criteria rubric (lasting impact, evidence quality, practical application, writing & originality, external consensus). Read our full methodology.

Leave a Reply

Your email address will not be published. Required fields are marked *