Hacker News new | past | comments | ask | show | jobs | submit login
Moonwalking with Einstein: The Art and Science of Remembering Everything (thegatesnotes.com)
98 points by rfreytag on Sept 16, 2012 | hide | past | web | favorite | 37 comments

I'd like to know more about remembering things that I don't put forth effort to remember. Does anyone know about this?

For example my mom asked who the angry chef on TV is and somehow I immediately knew "Gordon Ramsay!" even though I don't have cable TV, never purposely worked on remembering his name, and am generally terrible at remembering names. Being able to remember things without trying is what most folks would really want when considering improving their memory.

I have found success with methods that call for imagining a 3d space and placing objects in that space and using it to recall things... However, designing such a space takes time/effort which is difficult when you're trying to learn and comprehend stuff on the fly.

I do wonder if it is a memory issue or a recall issue here...perhaps I actually know the information but can't access it as necessary.

Even worse...

Daniel L. Schacter, in "The Seven Sins Of Memory" quotes Tatiana Cooley, US Memory Champion, as saying "I'm incredibly absent minded, I live by Post-its".

So the thing is ... trained recall doesn't necessarily help your "prospective memory", your "remembering to do things".

Your average technical person probably can recall a massive amount of stuff, based on association, about things that interest them. Trained memory works similarly and similarly has the problem that memories don't pop-up when you need them but rather when you make an effort.

The book touches on this. Basically, the human brain is biased towards novelty. It's much more likely to remember something if it's unique or interesting or different -- if it stands out.

The book's title, Moonwalking with Einstein, is one such example. Gordon Ramsey is probably another: there are lots of chefs out there, but if you see a chef screaming at people in his kitchen, it's probably going to stick with you, consciously or otherwise.

This is exactly my question. These techniques seem great for memorising (with effort) certain pieces of structured data that you know you'll need to recall later.

However, what about the things you don't realise you'll need to recall, or for things that you don't actively trying to memorise.

Basically - does it improve your 'base' memory, or are they just tricks you can use in certain cases.

Based on my understanding from the book, it's just a tool to memorize certain information. It wouldn't improve your everyday memory by much, if at all.

The memory grandmaster who trained Foer, Ed Cooke, runs a phenomenal startup that brings described techniques online to help you learn vocabulary: http://memrise.com

Highly recommended for anyone looking to expand their vocabulary in a variety of languages, including Mandarin.

This is probably the right crowd to ask: all the memory techniques that I have seen seem to be for memorizing sequences. Are there any techniques for remembering factual information more powerful than SRS? Like for the main points of a technical paper or similar?

This is in the QA:

Q: How is your memory now?

A: Ironically, not much better than when I started this whole journey. The techniques I learned, and used in the memory contest, are great for remembering structured information like shopping lists or phone numbers, but they don't improve any sort of underlying, generalizable memory ability. Unfortunately, I still misplace my car keys.

SRS seems to be the right answer for increasing the long term recall of just about anything. For main points of a tech paper, you could probably use the link method to associate a strong image for each of the main points in the technical paper and then link them.

Then you could pump those images into your SRS system for review.

You can find info on the web about link methods and recall, or you can just grab Higbee's "Your Memory : How It Works and How to Improve It" from Amazon. I found old copies pretty easily in university libraries as well.

You could represent those points as a sequence... it's basically the same techniques.

> When you're reading a book, you don't sit there and say, "Well, what does that word mean? What does that word mean?" After many sightings of the words in your vocabulary, your recall is immediate and very, very good.

But there's still ‘inner speech’, which appears to slow down meaning recall. It seems to be a habit of many people, although I don't have hard data to support this. I've heard that unlearning sub-vocal reading in favor of visual reading helps increase reading speed and remember the information. Haven't had much success with this myself so far, it needs some patience.

(I suppose phonetic writing system also adds to the problem. How cool it would be if you could read the meaning from symbols, instead of doing letter combination matching.)

>(I suppose phonetic writing system also adds to the problem. How cool it would be if you could read the meaning from symbols, instead of doing letter combination matching.)

Isn't this what visual readers already do, by recognizing whole words, or even groups of words at the same time? Observing myself reading, it seems like that's what I do.

Thanks for correcting, I'm not sure how I arrived at that point—it seems to me I too recognize word at a time. However, I'm not a visual reader as I understand the term: unless I consciously attempt to suppress my inner speech, I silently “pronounce” each word, which slows down the process.

I'm trying to imagine using a writing system in which how a word is written has no connection to its pronunciation, representing only meaning. That would pretty much make inner speech impossible, among other things.

"Never memorize something that you can look up." [Einstein]

I couldn't find a source that he said that. Wikipedia says it's not sourced, just a variant of:

> [I do not] carry such information in my mind since it is readily available in books. ...The value of a college education is not the learning of many facts but the training of the mind to think.

I haven't got the book yet but I'm planning to get it soon. As written in Q&A - The techniques I learned, and used in the memory contest, are great for remembering structured information - this might be great for learning languages. I remember grabbing a lot of new vocabulary by studying by heart lyrics of songs, quotes, jokes or whole sequences of dialogs from movies.

Just a head's up, this books reads more like a novel than a How-To. There are other books that are better for actually learning the memory techniques discussed in _Moonwalking_.

I don't mean to imply that _Moonwalking_ isn't worth buying; I think it's a really good book and was a very fun read, but at the same time, it was FAR less instructional than I had expected (though I did learn while reading it).

For a basic primer, I'd recommend this to get started with the memory palace technique: http://www.wikihow.com/Build-a-Memory-Palace

And for a much more comprehensive instruction, Higbee's "Your Memory and How to Train It" was the most referred to me: http://www.amazon.com/gp/product/B001Y35JI0/ref=docs-os-doi_...

This is presumably based on the theory that your mind has a certain amount of "free space" and that you don't want to fill that space up with a bunch of trivia. Anyone know whether cognitive science supports this theory? Bonus points for references.

I generally agree with this, but there are some things that are useful to have in your local cache for easy lookup.

unless you lose your internet connection...

Most of us will have to practice for months and months and months.

This isn't true. I've used this technique to memorize and recite the alphabet backwards. I also use it to memorize trig identities (I keep them in my silverware drawer, mentally). It's not "easy", but it only took one (entire) night of practice to recite the alphabet backwards. If I can, then almost anyone can.

Imagine the biggest house/apartment you've lived in. The only important criteria is that you've spent enough time there to have a clear, crisp picture of every room without physically being there or looking at photos.

Now look around. Take note of each "container" -- bins, cabinets, drawers, and even furniture like chairs, tables, and couches (because you can set things on them).

Memorize a specific path through your house. It sounds hard when I phrase it like that, but it's not. Just start at your front door and mentally walk through your house, opening cabinets/drawers and visiting rooms in a specific order. The order is all that matters.

As you go along, imagine what it would look like if there were literally an alphabet in your house. Starting from the front door, a big red hairy "A" is blocking your path, so you go around it. Then as you pass your hallway closet, you notice a shiny blue "B" isn't doing a very good job of being inconspicuous. And so on. (It helped me to make them distinctive, so I use a lot of color and tactile imagery like hairy/furry/velvety/shiny. Even smelly. Just use your imagination.)

Remember -- you have to make sure you always are traversing in a specific order. The room with the "A" to the room with the "B" etc.

Since most of us don't have 26 different rooms, you'll have to use your containers too, and also cram a few letters together into one area. (Never more than three in a bin, though, unless you're better at this than I am.) So in your livingroom, maybe S T and U are siting on your couch watching The Office, whereas X Y and Z are getting freaky in your bedroom. (Shocking imagery is easier to remember than the mundane, and you can use this to your advantage.)

So. Got your path memorized? A to B to C etc...? Good. Now it's very easy: walk the path backwards. Z to Y to X...

Speak the letters out loud as you go, and presto, you now possess a new (entirely useless) talent.

It becomes much more useful for pretty much anything that benefits from rote memorization. (Just don't apply it to math very much otherwise you won't have more than a superficial understanding of the underlying concepts and connections.) It's quite handy for presentations.

I find that creating a simple song is easier. I sing the alphabet backwards, in the same tune as the original.

I read the book. It's a great read for the story itself (guy learns how to win memory contests) ... but the memory techniques, in my opinion, do not really work for everyday situations. I did learn how to memorize the sequence of a random deck of cards though, which is pretty cool.

I enjoyed the book as well. How long did it take you to memorize the deck of cards?

Took me a few weeks ... although I carried a deck around with me and pulled it out to practice periodically throughout the day. I found that the key is to not be afraid to be really creative with imaging the visual representations of the cards. The more 'out there', the better.

Josh Foer gave a good TED talk about this topic, you can watch it at


After seeing this I actually was able to use his technique to help remember all the parts of a long wedding toast I had to deliver. It definitely worked in this situation, though I'm not sure how applicable it would be for cataloguing every day information you may or may not need to recall.

If anyone is interested in the historical development of these practices (as Gates mentioned), they are known (in Latin) as ars memoria. They were extremely common in ancient Greece, but (ostensibly) came under attack the the development of writing (Plato complained bitterly about it). They were revivified in the Renaissance, and were connected to a whole host of linguistic studies, including the development of "universal" or "artificial" languages, and, interestingly, cryptography.

short summary from the book in order to recall sequenced information:

walk down a path in your mind, that you can recall well from your memory and then place images of the items/words you try to remember along this path.

in general: the more numerous the set of associations you have with a certain word the more likely you are going to remember it. For example, when trying to remember the name Baker of a person, don't think of it as a surname but rather as the profession "baker".

I read this book well over a year ago, and I still remember nearly the entire list of items he used as an example in the book. The techniques described really work, though the book is more of a memoir than a guide.

Here's what I remember, for the record: pickled onions, cottage cheese, peat smoked salmon, six bottles of wine, three pairs of socks, hula hoop, scuba diver, dry ice(?), Sofia Lauren.

I know there are a few more, but I think I covered a pretty considerable number of them.

Ope. Looks like I was actually off by a fair bit. :P Here's the actual list, from http://www.goodreads.com/review/show/155631525:

Pickled garlic. Cottage cheese. Peat-smoked salmon. 6 bottles of white wine. 3 pairs of socks. 3 hula hoops. Snorkel. Dry ice machine. Email Sophia. Skin-toned cat suit. Paul Newman film. Elk sausages. Megaphone and director's chair. Harness and ropes. Barometer.

I read the book earlier today, and I find it somewhat bizarre that I can rattle off the list too. (With no errors, but I only just read the book.)

NPR audio interview with the author and journalist Joshua Foer: http://www.npr.org/2011/02/23/134003962/Moonwalking-With-Ein...

I read this yesterday, got inspired, and memorized 50 digits of pi starting at 11 pm. Crazy! It's really not that hard!

I went to one of his book talks, and it was really exciting to hear about. I think he mentioned how it was a one-time rise to stardom, and put me onto the glacial blog[1] that is Climb For Memory, by the guy who has won it for the past couple years.

[1] http://climbformemory.com/blog/ Oh, it's moved, I guess that explains that.

Really enjoyed reading this book some months ago. Certainly worth some of your time.

The ancient Indians perfected memory retention of large bodies of knowledge (called Vedas dating back more than 4000-6000 years) with the help of very elaborate schemes. The Vedas are composed in a language called Vedic Sanskrit (althought similar to Classical Sanskrit, still distinct). These schemes not only aided in retention, but also preserved the phonetic purity and semantics. The syllables were also intonated. Loosely speaking (the reason why I say that is I still have not found a compact and crystal clear way to convey what Vedic intonations are there for), intonations are also representative of grammatical case/context. So, the syllable intonation also had to be maintained. A brief preview of some commonly employed chanting schemes are given below:

For describing the various schemes let us consider a sentence with five words having indices 1 2 3 4 5.

Let !! - represent a termination sequence. We'll not get into technicalities, but take it as is.

1) Krama Paatha - This is relatively easy and each of the indexed word will be chanted as follows.

1 2

2 3

3 4

4 5


Notice that a given word is clubbed with the next word. (Forward chaining). There are two words in a sequence

2) Jataa Paatha - This is bit more complicated since it involves backward sequences as well.

1 2 2 1 1 2

2 3 3 2 2 3

3 4 4 3 3 4

4 5 5 4 4 5


The complexity is due to the fact that the words will have to be chanted in reverse sequence as well. (backward chaining) For e.g. {1, 2} {2, 1} {1, 2}. There are a total of 6 words in each sequence.

3) Ghana Paatha - This is the most complex chanting mode (also called the bell mode of chanting). The chanting sequence is as below.

1 2 2 1 1 2 3 3 2 1 1 2 3

2 3 3 2 2 3 4 4 3 2 2 3 4

3 4 4 3 3 4 5 5 4 3 3 4 5

4 5 5 4 4 5


Phew!! Each sequence has 13 words made from three words. There are two word forward/backward sequences as well as three word forward/backward sequences. For e.g. {1, 2} {2, 1} for two word sequence. {1, 2, 3} and {3, 2, 1} for three word sequence. The astute reader will observe that the penultimate sequence is the same as Jathaa Paatha (and also the first six words of any sequence constitute the same).

There are several intracacies that are involved. One is compounding of words. Word compounding introduces changes to the pronunciation. Another is the change of phonetic intonation due to the modified sequence of chanting the words in a given scheme. These repetition techniques enabled the scholars to get a "feel" for how words would be intonated and pronounced phonetically in various circumstances.

I had always wondered why such elaborate schemes were devised by the Vedic seers. I have come to realise that the seers knew that when knowledge was transmitted in the oral form, it inevitably introduced corruptions in the words. In Vedic Sanskrit, purity of syllable (both phonetic pronunciation and intonation) is of utmost importance. Even a slight change in a syllable (either phonetic or intonation) would completely change the meaning of the sentence.

These techniques not only aided in error detection in chanting, but also error correction (specifically error correction in syllable intonation). If someone were chanting a sequence of words, one could verify if the words are being intonated correctly (of course it would take an expert to do that). There was a document that I read quite some time back which resorted to some abstract algebra (group theory) based arguments for explaining the error correction mechanism of some of these techniques.

Of course, many of these schemes (and also other linguistic aspects like word compounding) always kept one thing in mind - the ease with which humans are able to cognize and generate the sounds representing given syllables in a sequence of words.

In conclusion, these methods ultimately enabled scholars/priests to be able to: a) remember entire bodies of verses/compositions with relative ease. b) transmit such knowledge to students with zero word corruption (yes, although I am neither a linguistic nor a Vedic expert, I can go out on a limb to make this claim - zero corruption).

Edit :- Minor edits for readability/spelling.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact