Hacker News new | past | comments | ask | show | jobs | submit login
"Memory athletes" and the techniques they use [audio] (bbc.co.uk)
111 points by open-source-ux 31 days ago | hide | past | favorite | 50 comments

Eidetic memory is one of those things only pre-pubescent kids are supposed to have, and mine survived into high school. I can still recall a couple history tests where I could not remember the answer but I knew for sure I wrote it in the margin on a page of my notes. When I finished the rest of the test I ran the clock down racking my brain trying to tease out the exact text.

Because of this I always thought of myself as a visual thinker. Until the last time we were discussing Aphantasia here and I realized to my horror that I hardly ever picture anything in my head anymore. Which is super confusing considering that while I'm awful with names, I will remember your face pretty much forever. We've met before, I'm sure of it. I am so assured of this, that on the rare occasion someone recognizes me but I don't know them, it's so unsettling that it borders on traumatic. Like the guy in the movie who discovers he's been spied upon.

What I've worked out is that I'm a spatial thinker, which goes to explain why I get very particular about the 'shapes' of code the way some craftsman get snippy about someone moving their tools. Once I teased that out I then discovered that a lot of my ways of processing complex data are functionally indistinguishable from a Mind Palace, but the decor doesn't matter, just the shapes. Like navigating your house at night without turning all the lights on. Careful, there's a chair there, and the dog bowl is over here. And at the store, this couch is great but it won't quite fit in our living room. Yes, I'm sure honey.

Wow. You just made me realize that I don't (can't?) do that anymore. I used to have the exact same ability though high school. I would remember where on the page I read something or wrote a note, then during English exams I would find a quote by flipping through pages while looking only on the spot of each page I expected the quote to be on.

I can't really picture things in my head as an image, yet I think very spatially. For me it is not an image, but shapes that I can feel. I design 3d models by feeling them in my mind, where I can feel my way around the model to find problems with my design etc.

I was going to respond to another responder that it feels like reaching up to adjust the shower head when you’re facing away, but that’s not entirely right. It’s more like... this is going to sound weird, but like it’s rotated in a higher dimension. It’s made me curious about 4D VR simulators. I had this feeling that perhaps they’d make a sort of sense to me, but the ones I’ve seen so far don’t, and I don’t know if it’s because I’m wrong, or they are.

That “feel” extends to fitting things together. I had a boss who would ask to fit something else into a UI, and every time I told him it won’t fit, he didn’t believe me. So then we had to waste a man-week to show him he was wrong, and then he would forget that I told him that already and we’d do it again in a few weeks.

Man I was so glad when he laid me off. Best severance I ever got.

If you had eidetic memory, you would know exactly what you had wrote in said margin.

Eidetic memory comes in grades. It usually seems to make use of or abuse spatial memory to remember facts in great detail that you then synthesize into semantic information. See "memory palace."

Eidetic memory used to be considered a negative mutation in that the bearer was not "properly using" their semantic memory, i.e. just remembering the fact without having to synthesize it from a raw sense memory.

From Wikipedia on eidetic memory:

> To constitute photographic or eidetic memory, the visual recall must persist without the use of mnemonics, expert talent, or other cognitive strategies. Various cases have been reported that rely on such skills and are erroneously attributed to photographic memory.

Eidetic memory is exceedingly rare more on the "single digit adults have ever had it", many scientists don't believe it is a thing. However, you also have things like:

> von Neumann also possessed the power of absolute recall. Once he was asked to recite “A Tale of Two Cities” before being asked to stop after a perfect 15-minute recitation.

> John von Neumann was able to memorize a column of the phone book at a single glance.

This is on a whole other scale than what the original poster said.

True, but note it is consistent with what I gave. There is more to memory and the study of the mind than Wikipedia. There are people with very good recall that do not fit that definition, or who recall things in a different way, but share similarities.

No, you said eidetic memory comes in grades. It does not, it is a binary ability the same as perfect pitch. Obviously there is more to memory than an ultra rare ability, but that was never part of this discussion.

> eidetic ī-dĕt′ĭk adj. Of, relating to, or marked by extraordinarily detailed and vivid recall of visual images.

There are grades of "extreme detail" ranging from instantaneous recall of your lunch three years ago as well as what everyone was wearing and what you were talking about to very very well above average recall of useful information in the form of sensory memory.

The justification for this interpretation is on Wikipedia and is related to the physiological explanation of eidetic memory. Instead of semantic memory sensory memory is used, and you can see this occur to various degrees in people.

The qualifiers are loose and mean different things to different people. You are not sovereign of the English language. The second definition given in my source matches yours, but it is not the only one.

The English language isn't sovereign to scientific terms, the science community is. If you pulled out a dictionary definition of "policy" in the reinforcement learning community, you would be ridiculed. These terms have clear and agreed upon definitions.

Etymology or dictionary definition is not 1:1 with a scientific term.

I'm similar. My memory is very dependent on spatial reasoning and I've sort of innately created strategies as I was growing up while my photographic memory was losing its strength.

For example, I remember linear information, like dates, by instinctually visualizing circles within circles. A circle for the years, circle for months, then days, etc. with specific times as sections of the circle. There's no particular reason for where things are on the circle though. For example, the month of May is at 12 o'clock or at the top of the circle while January is at 7 o'clock. For years, the current year is always at 12 o'clock. Same with the current day of the month, with weeks being sections of the day circle.

Hours and minutes, on the other hand, is like Penrose steps with each of the four sets of stairs being quarters of the day that I travel on as the day continues.

I think you just described my brain. Are there resources around spatial thinking you find insightful?

I had mostly developed my own systems/coping mechanisms by the time I figured this out, so I didn't really go looking for external advice that much.

My main way of absorbing material is making hypotheses and knocking them down until I can't. Remembering a model is easier than remembering unconnected facts. But I'm not sure if that's more to do with the spatial thinking or the not getting advice half of that equation.

You might enjoy mind palaces, mind maps, graph theory, "Refactoring" by Martin Fowler - though I can't entirely explain why, and activities that demand proprioception (martial arts, dance, rock climbing).

john von neumann had it his whole life

I like to think such people are actually normal, and we all suffer from some kind of brain damage that we could fix somehow.

"Moonwalking with Einstein" by Joshua Foer is an interesting book on the subject of memory athletes. The author is a journalist who sets out to write the book and in parallel practices alongside his research subjects. In the end, he winds up winning the USA Memory Championship and setting a US record. Really drives home the point that it's all about the techniques and the time one is willing to put into them.

I read it in 2011 when it came out, and I still have a vague memory of the memory palace he teaches you to build. I used my childhood home to remember:

- Garlic in the driveway

- Six bottles of wine playing catch with three pairs of socks

- A scuba diver in the kitchen sink

- A smoke machine in the living room

I have no idea what it was that I was supposed to be remembering or whether those are even the right things, but memory palaces are apparently a powerful tool.

The problem I've had with memory palaces is that I find it hard to keep multiple versions of a scene separate from each other. I also find it tedious to come up with the scenes themselves. Mnemonics are the difficult part.

Derren Browne explains as way to memorize a deck of playing cards by assigning mnemonics to cards and creating a story with them. I found it somewhat ease. Memorising the cards by themselves is tedious. "Jack of diamonds... wait, no, was it hearts?" There are too many similar cards. So you turn each card into a 100% unique mnemonic using a reversible method. He used a combination of the peg numerical system (1=l, 2=n, 3=m etc) and the first letter of the suit (h=heart, s=spade etc) to come up with a single meme for each card. For picture cards, use characters or people who match up. Queen of hearts might be your wife. King of clubs might be a famous MC or rapper. This gives you 52 items like:

Queen Latifa: queen of clubs

Ham: 3 of hearts

Elon Musk: Jack of diamonds

Car: 4 of clubs

Dog: 9 of diamonds

All you need to do, as you're memorising the order of the deck, is to create a narrative around the individual cards. "Queen Latifa is eating ham with Elon Musk in her car. Her dog is on her lap begging". There's no way you're going to get the cards wrong, or the order. The first time I ever tried it, it took me about 30 minutes to invent or memorise existing card pictures. Once I tried to use it I successfully remembered 25 cards in a row. Surprisingly, I could recite them backwards or name any card that came before or after any card chosen from the deck. Our brains are very good at remembering stories.

It's funny, the memory palace idea always seemed like such an odd concept to me, until I realized I have aphantasia and lack the "normal" visualization capabilities.

I have heard people who lack strong visualization are still able to get some utility out of it by verbally elaboratively encoding the thing rather than visually encoding it.

Totally. I did something along those lines as a kid to memorize the first 30 digits of pi. Basically coming up with an elaborate story and committing it to memory.

>coming up with an elaborate story and committing it to memory

I've tried a few of these, including the one in Derren's book. I have to say, none of it worked. Over the course of trying to remember 20 places of pi by various methods, I ended up committing it to memory too, but, and here's the important part, I do not remember them now. I think remembering different things benefit from using different techniques, and we just use the right one unconsciously, there's not a single method you can impose on all memory tasks.

same. I remember the bottles of wine, scuba diver and claudia schiffer swimming in cottage cheese or something like that. Socks on the living room couch?

Highly recommend the book, its a great read.

it is a great story and probably even a better book, but the sciences is really lacking https://greyenlightenment.com/2018/08/14/bullshitting-with-e...

There are no peer-reviewed studies that corroborate the claims of the book, and research shows memory champions all have Mensa-level IQs. This is not surprising. Short-term memory is g-loaded and is a component of IQ test (it is called digit span or digit recall), so it is logical to assume that the best at memory would also have high IQs.

That is a rather unconvincing post. It makes the link to IQ without any actual evidence, claims Foer has an IQ of at least 130 again without anything backing it up, and trots out a memory athlete's opinions as if they were unbiased facts. Never mind that the said expert specifically pointed out that correlation is not equivalent to causation. Or that IQs as measure of intelligence is still pretty contested.

It's been a while, but I have a recollection of Foer going into the question of why more people don't do this in his book. It mostly has to do with the amount of time one needs to commit to practice, coupled with limited transferability of this skill.

To give another anecdote, I've worked with a very high-performing engineer who spent a bunch of his free time building memory palaces for new codebases to ensure retention. I'm sure he was above average IQ-wise, but his productivity likely benefited from his approach as much as from his innate abilities.

It is more than unconvincing. It is outright misleading. The idea that scientific investigations are refuting the idea that these skills are learnable is false. It doesn't make much effort to find studies to the contrary. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4056179/

It also goes against common sense. Math, reading, and so many things are all learned, but frequently used by people who have a high IQ. Are we to assume that these things are not learned, because the author thinks IQ is the cause of reading skill, math, and similar things?

I've bothered going through the labor to learn the skill. I've seen my baseline measurement of recall ability improved by the training time I invested in learning the skill. I haven't bothered to get to a competitive level, but I don't need to do so to be certain that humans can learn skills by practicing them with the intention to improve.

The skeptic is just wrong. People can learn to improve their memory.

Consider that people with higher IQs have greater proficiency at math and reading (such as SAT scores) despite everyone regardless of IQ having some ability.

Yes, but that doesn't reflect the inability to learn. You're confusing cumulative advantage leading to a competitive edge with an inability of lower IQ humans to learn advanced skills. You will also find things like richness and beauty positively associated with winning competitions too. That you will find this doesn't mean memory can't be improved. It reflect cumulative advantage more generally. It is one of the reasons for the cognitive bias of the halo effect, a statistical side effect of the principles underlying the Matthew Effect. Edges are edges.

Critically, that doesn't mean that the skills aren't learnable. It means that people with the most advantages can exploit those advantages to gain a greater deal of advantage than someone who has less advantage.

And so what if success in memory competitions is correlated with IQ? Grades among students are highly correlated to IQ, but that doesn't mean that good study habits and hard work won't improve your grade. Though I don't have a peer-reviewed source to back that up, sorry.

The article is wrong in many places. For example, it says:

> –One study showed that that training is non-transferable. This means if one learns a sequence of numbers, the skill fails when one tries to recite them backwards or a new set of numbers.

but just about every person who has ever memorized numbers with those memory techniques can recite the numbers backwards too. If you gain the skills to memorize one page of random numbers you can memorize a different one too, no problem.

On the other hand we have the Savant syndrome which is typically not associated with high IQ.

That, and I'm glad I didn't read "Mensa-level IQs" while drinking coffee, I would have spit all over my keyboard.

Mensa? A low threshold, yet smart enough to be in the game if you can find something that really interests you. But you choose a club? Huh...

It's actually heartening that various people famous for their IQs have accomplished zilch in their lives other than being famous for their IQs. This tells you that the test is flawed, and that other factors determine revolutionary levels of success.

The magician/mathematician Persi Diaconis long ago explained to me about "memory palaces" and their effectiveness in competition. This can be generalized: The mountaineer Conrad Anker relays how European skiers can spot an American skier from far off: They waste effort. Many people had Michael Jordan's physique, but Michael Jordan spent a decade developing his physique into a world class athlete.

One is the architect of one's intelligence. The most poisonous aspect of IQ tests is that people believe them, and conclude that intelligence is innate, not malleable.

If you think Mensa people accomplish little, wait until you see the general population

I'm going to hang this bit of trivia here ..

Robert Forster (of "Jackie Brown", "Breaking Bad" and much earlier "Medium Cool") was a member of the high-IQ Triple Nine Society.

This is an excellent 30 minute radio programme/podcast. Since most people I suspect won't have 30 minutes to listen to it all, the programme finishes on an interesting note.

There is a cultural stereotype that Chinese schooling places a higher value on memorisation of facts, compared to liberal Western education systems which prioritise understanding over rote learning. So these memorisation techniques and competitions are painted as a difference in education values.

But as one of the "memory athletes" says, the techniques of memory training are not rote learning, but inherently creative. He goes on to explain why.

There's much more in the programme including how to create a "memory palace".

The most important thing I learned about this sort of memorizing is that it is not useful to me, and that let me stop worrying about it. Sure, there might be some cases where it could be, but for the most part this technique overfits to the competition and doesn't transfer to my day-to-day.

For things I want to memorize day to day I don't have to memorize a long list of things and recall it directly, I might have to build up a list of things I can recall, but I have time to build up that list, and I can use spaced repetition (Anki) for that.

For those interested https://artofmemory.com/ is a forum where lots of memory athletes share strategies and tactics. Additionally it also has some software you can practice to improve your skills.

I took a "memory course" in college and it was like a muscle based exercise. If I used it and practiced it, it worked. But after I stopped practicing/using it for a while, eventually it became atrophied and I could remember things as well as I could when I was working hard it.

Is this how it works for everyone that "develops" a good memory?

I think this is one of the main points in moonwalking with Einstein, discussed in a sister thread.

By the end of the book, the author wins the American memory championship and the following day he forgot where he parked his car.

These memory techniques are like a notebook. They won’t make you remember stuff any better, but if you wrote something down (in your memory palace or notebook) you can look it up.

For storing random key value definitions, mine seems to work that way. But it kind of helps me to think of memory as a network of interconnected bits. One random point will be hard to get back to, but if you have a dozen related bits of information that all link to one bit, it will be easy to find your way back to it and recall it. Building a solid base of knowledge and understanding of a subject helps to build these interconnected networks.

Can't remember.

I once met Jim Karol who uses recursive associative mnemonics to form a n-dimensional associative memory matrix. Most people have a 1-D or 2-D associative matrix, but Jim Karol can basically store associative memory with log(n) associative lookups.

Nice. Can you explain the recursive aspect? A quick Google reveals only teaser articles and ads for books. I'm interested because years ago I started developing a "nested" system, although I never got it adequately fleshed out.

For the curious, mine was about dates. 31 words for days in the month, 12 for months. The secret sauce, though, is practice. I never did the practice, so I remain limited to the first three dates I tried to commit to memory with my system. But: I can still remember them! :)

I kind of do that sometimes with little framed vignettes leading to other memory palaces inside stations of memory palaces.

If you want to watch top memory athletes compete in real time, check out https://memoryleague.com/

You can also watch the ongoing championships on Twitch: https://www.twitch.tv/memorysportstv

One of my pandemic hobbies was learning all the days of the week for a few years. Actually pretty doable.

what does that mean?

Presumably, given a particular date in the future, telling you what day of the week it would be. However, there's also a method you can learn to help you accomplish the same thing, without the need for committing a great deal of facts to memory: https://en.wikipedia.org/wiki/Doomsday_rule.

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