Method of loci criticisms

General Memory Chat
1

EDIT: after reading this post, make sure you ready my last comment.

After experimenting for a few days with the method of loci, visual linking and pegs, I think the method of loci is a poor choice for everyday use. The method of loci’s main advantages over using pegs or visual linking, is that you can jump to any room, so you don’t forget the rest of the elements if one item cannot be recalled, and that it minimizes interference, as long as rooms have enough contrast between each other.

Regarding losing the rest of the items if the chain is broken, for everyday use, people don’t even learn hundreds of items, they learn dozens at most( including med students). You can still forget items with the method of loci, it’s just that the damage is diminished; but the damage is diminished with shorter chains already, and you’ll have pretty good recall just with pegs or visual linking. Thus the advantage the method of loci has when using small chains, is negligible.

About the minimization of interference, very very few everyday things are prone to the interference the method of loci helps with. When you are memorizing decks of cards, the method of loci helps because this deck of cards is just like the previous and the next one; just a deck of cards. You could assign a number to your attempts, or a mental image if you like, but I doubt any mnemonist has tried that, as the method of loci works just fine for this. The reason the method of loci is not required for everyday things, is that if you are, for example, memorizing the list of symptoms caused by X condition, there’s only one list for that condition; another condition will have a different list, and two medical conditions will never have the same name.

Thus, people who are not aiming to compete, should be directed to linking and pegs, NOT the method of loci, because their time and effort is better spent on those techniques as they require much less resources, will give the same result, and more people will be enthusiastic about learning more about mnemonics.

Mixing the method of loci with pegs or visual linking for everyday use is not the answer either. It gives you the disadvantages of the method of loci, without any extra benefits.

There is one exception though: repeated temporary storage. The method of loci can then be used with pre-made rooms, that are overwritten, therefore reducing the cost of the method of loci to a single one time big cost, and the minimization of interference is actually useful in this case. This can even be useful for long lists when you cannot write or type down the information, as long as you remember to dump the information later to clear the palace for later use.

You might say that the problem is that I’m a beginner with mnemonics, but I hypothesize that the cost of creating new rooms using the method of loci for non-temporary storage will increase when you run out of easily accessible palaces to use; so the cost creating new rooms with the method of loci is NEVER equal or less than the cost of pegs or visual linking, at non-beginner levels. I’m not sure if the cost of pegs or visual linking recall is smaller than the method of loci though, because the method of loci givesa visual context, which might be more efficient than no context at all.

Opinions?

1 Like
2

No.

I recommend using method of loci combined with linking(/pegging/stories), as the optimal length for a chain of links(for me and others) seems to be 3 to 5. After that, the chance of breaking the chain becomes much higher. This way, you don’t need that many loci.

I really do not understand why new people(and some others) always talk about the “cost” of creating new palaces etc. It’s really not that hard. Go walk around a park. There you go, 30-50 loci. 15 minutes. Take pictures if you forget locations easier than most. You have many locations already memorized. Ever play 3d video games? There you go, you have 20 “maps” with 50-100 loci in them. You can even use movies. If you don’t like leaving the house, use google street view, or stuff like this to explore locations. To practice those journeys, memorize 100 numbers in them(or however many it takes to fill the whole thing up).

Once you memorize two or three number strings in a particular journey/memory palace, you’ll know where the loci are 99% of the time. There isn’t much danger of missing them.

It’s best to combine the two methods, put things in locations AND link them into stories. It’s very memorable that way if you do it correctly, and it’s not that hard.

Bateman

Edit: It’s also much much more organized haptic link chains. You can store all your notes on physics in one particular memory palace, you can store the main points from a book on body language in another, and you know where they are and can very easily access them.

2 Likes
3

100 loci is not that much; med students are regulary given more than 10 pages daily to learn, and I’m pretty sure you’d need more than 100 loci to accomodate 10 pages. Of course, you shouldn’t use mnemonics for everything. Not everyone approaches mnemonics to achieve great feats, but just to learn faster. If newcomers repeatedly mention the cost, it should light up a bulb and tell you it IS an issue.

I have no problem with the amount of loci however; I easily picked a palace with plenty of available space, however, besides places that I frequent (home, workplace), I can’t instantly visualize a vivid image, I have to work out the details a bit, and make sure rooms won’t get confused in my mind, as it has already happened to me after just using 4 rooms; and you have all kind of small things to fine tune: how detailed should the room be? how many items can I fit in a single loci? etc… Then I have to walk through. It might also be that some people might be better suited for the method of loci than others.

Practice requires plenty of free time, and newcomers don’t even know if mnemonics will work for them, they are trying arcane territory.

My point is to make it clear that the method of loci is more costly that other methods, at least for beginners, even when they have selected a palace of a convenient size. Ignoring this, and putting blame on newcomers by saying they are lazy is just going to put people off mnemonics.

Extra: I did try number strings with the method of loci, and it works fine. However, for me images are more difficult to maintain, which is funny because images are easier to learn the first time. It is possible that I’m “doing it wrong”, because I am using linked lists within my loci, so that might be too complex for a beginner, or because my images are often complex (I suspect simpler images might fare better), but this just supports my point that the method of loci requires more fine tuning than other techniques, as my images work just fine with visual linking.

4

Depending on how detailed the pages are, how much information you already know/can easily memorize using natural memory, and adding on links, one could store 300-400 images in 100 loci, which should be enough for 10 pages. But yeah, med students have a lot to learn. Conversely, solely using the linking method one would need 100 linking chains organized in some form if they linked 3-4 items together. The more links you add on, the higher the chance of the chain breaking, but 3-4 usually doesn’t. It’s much less efficient.

Yeah, I usually walk through a new journey beforehand, pick where the loci will go and then fill it with some random info to make the loci stick. Practicing will make it a lot quicker, and improve your visualization skills.

As you’ve said, mnemonics aren’t suited for every single thing you might want to memorize.

You’re probably not concentrating on putting the images in the background of their loci when using memory palaces. You have to make it connect with the locus. This will probably eliminate a lot of the forgetting.

I didn’t mean to imply that they were lazy, perhaps just unaware of the variety of things that can constitute memory palaces/journeys, and the ease in which you can create them.

I would again recommend even if you just want to link things to always use a locus as the base for the linking chain. This will help you find it, and to not lose the beginning of the chain.

If you only need to memorize small things like grocery lists/phone numbers, you don’t need a huge memory palace. Obviously don’t use it if you don’t want to.

Bateman

Edit: You should check out gavino’s massive memory palace system and r30’s post on using movie scenes and also rotating scenes to very quickly get many more loci. Perhaps that would make it easier.

5

I haven’t had trouble linking more than 3-4 items, and even though there are more links, they are pretty cheap in terms of time invested. If you refine the method of loci, it seems it would be cheaper, but for that you have to go through that refinement phase.

I did fail to put the images in the background of the loci at first, but that was fixed a few days ago.

I know you didn’t mean any harm, but when you say “you just need to put more effort” it can be interpreted in that way.

Thanks for your replies!

6

To clarify: Links are “cheap” in terms of time investment, but when combined with the method of loci(which takes a little time to think of a place and get it ready) are far more memorable and effective.

Glad we came to an agreement of sorts.

Seems like I responded to every single one of your posts so far :stuck_out_tongue:

Good luck,

Bateman

7

It is possible that the bonuses provided by the method of loci are not necessary for everyday uses, no need to use a nuke against 1 person; additionally, I know at least one person has won the memory championship just with linking, which proves they are quite reliable too.

EDIT: after further practice and research, it seems that the method of loci does have an advantage over linking; the spatial memory, while it works together with visual memory, stores information in different areas of the brain. While it might be harder to process information from multiple senses (that is, giving something a smell, a sound, etc., besides just giving it a visual form), the more senses you use, the more neurons your information will be related with, and the easier it will be to store in the long term memory, specially since it seems that purely sensory information is stored in the long term memory without having to go through short term memory, together with emotions.

Therefore the method of loci is more likely to form long term memories than linking or pegs on their own. This doesn’t invalidates the rest of the points I wrote about before, like interference.

8

The technique is much less interesting than the results, something that can only be decided by the user. To discount any method or make a universal statement is folly.

That said, as Coz concludes, logic cannot help but dictate that “loci” (outdated term) is more likely to form long term memories.

But that again is irrelevant.

Why?

Because, with the exception of competition (which very few people engage in and even fewer amongst these need higher-order skills to succeed), few memory techniques have a built in structure upon to engage in no-brainer repetition cycles that allow you to overcome the forgetting curve and literally defeat the serial positioning effect by virtue of being able to leap frog around.

If there were a superior memory technique that will serve most normal needs, it would have to be Memory Palaces used in combination with associative-imagery which almost by default involve some kind of linking.

Thus, the question is not how to separate the techniques and place them in a hierarchy. The question is how to bring them together and teach them to others as an interactive toolbox, knowing that you can use most techniques inside of Memory Palaces, but usually not the other way around.

9

Why would you say loci is an outdated term? all it means is locations.

My concern is (now to a lesser degree), that the method of loci is much more complex than other methods, but it’s the first recommendation given to beginners, even if the majority of the benefits of mnemonics can be obtained with simpler techniques. It could be that due to the usefulness of the technique, and how easy it is at the expert level, that what beginners say might be disregarded.

10

Thanks for this, Coz.

Perhaps “outdated” is the wrong word, but as someone who teaches memory skills every day - but still loves obscure words and etymology and all manner of hapax legomenon - I stopped using it long ago. Partly because the term constantly requires two things:

  1. Explanation of what it means with some kind of history

  2. Pronunciation training

Neither of these things are bad in and of themselves, but neither involve a direct line to understanding the methods. Right or wrong, some people also find unfamiliar terms alienating and they are turned off the practice right away as soon as they see it.

Plus, “loci” often isn’t specific enough, certainly not for many beginners.

What I’ve done to replace it is probably detailed to a fault. I only teach Memory Palaces and if the word “location” is used, it means a Memory Palace based upon a building. The journey through the building has “stations” where associative-imagery is placed or stored (also true of an outside journey).

Then there is the difference between macro and micro stations, entire rooms or the use of elements inside of rooms respectively.

Again, “loci” just isn’t specific enough and in coming up with detailed terms (and I have many more in my teaching), I help far more people on a daily basis than I would by sticking with “loci” and then unpacking it again and again with a different story every time.

And since mnemonics are both scientifically verifiable and even a scientific pursuit in and of themselves, I think the art of memory can and should evolve to have a more precise glossary.

In sum, Memory Palaces premised upon “loci” are not complex and when used well, their benefits exceed all other mnemonics by far because all of them can be used inside of Memory Palaces.

Not only that, but with proper guidance, Memory Palaces couldn’t be easier to use for the majority of people once you’ve explained them in solid terms.

Anyhow, these are just some of the things I do and being detailed, systematic and clear in the terminology by eschewing loci in favor of more specific and detailed explanations (except for the history buffs and nascent memory nerds, whom I always send here) gets amazing results for people.

And results are what matter.

11

I still don’t see how it is “much more complex”. In linking, you turn ideas into images and combine them, either by transformation, action, creating a story, etc. In memory palace you turn ideas into images, and you combine them with wherever you want to put them. Sure it’s slightly more difficult since you have to think of a place whereas linking is just in limbo. My concern isn’t so much with beginners, it’s with people who haven’t tried the techniques to any real degree trying to explain them. I’m sorry if that’s not you, but that’s the impression I get.

I’m glad you researched and practiced more about this, and thus changed your opinions and subsequently your comments.

As Metivier said, it’s not about separating the methods, it’s about using them to their fullest extent. It’s not so much which one is better, it’s where are they most effective. I also agree with Metivier on the term “loci” being unnecessary. Much easier to just say place, location or as Metivier does “station”.

Bateman

12

Hi Coz, this is an interesting subject you brought up as it does seem like you dived head first into this without years of experience behind these techniques. I agree with Bateman and Metivier, I think there is a vast range of people who can give little insights into developing what we take for granted.
I’m also new to memory techniques and would say I use method of loci, but in truth I use as many memory techniques I know about to support the method of loci. To try and divide them up, seems like saying the heart is better than the lungs when as a Med student you know they both need each other to get the desired result of transporting oxygen and carbon dioxide.
I think if you can give an example of what info you would memorise and how you use linking and then method of loci for storing that info. Then maybe we can get an understanding of how you process that information into your memory. Each persons mind is unique and maybe you put the info together in a way that just is more suited for linking. Or people can give pointers on making the most of Method of loci.

13

I have gone further into the method of loci, and I believe I use it, though maybe not in the traditional sense; I figured that the advantages of the method of loci are triggered, not by picturing a 3D space, but by propioception, which means, you have to imagine yourself seeing face-to-face the objects, or even better, doing something with them.

Every introduction to the method of loci I have read starts with the task of creating a memory palace; instead of pre-creating a memory palace, I create a location on the spot. One of my favorite ones that I recently created is me ‘standing’ on the side of a pine tree, looking down on a man, as he morphs through several stages, each one representing an item in a list. Some of the my loci are based on reality, some are completely artificial.

The main problem I’m having at the moment, is that sometimes I don’t connect the key with it’s mnemonic. For example, if I am asked which are the factor associated with X, I simply might not remember the visual image at all. I have read that when this happens, the most likely explanation is that your mental image is not shocking enough. However, I’d say that I remember 7 out of 10 things I use mnemonics for, much better than the 4 out of 10 I had when I wrote this post.

1 Like
14
I figured that the advantages of the method of loci are triggered, not by picturing a 3D space, but by propioception, which means, you have to imagine yourself seeing face-to-face the objects, or even better, doing something with them.

This is a very poignant observation. We have (at least) four reliable stores of memory in our brain: spatial, visual, muscular and aural. By using movement and action in your memory palace, you are encoding onto your muscle memory which will definitely strengthen the memory.

I’ve been using actions to unlock embedded memory palaces at my locations and it works very well. Now I know why. Thanks!

15

i am about to give up competitions of memory for now.
I will share all my experience about memory sports
i was unable to stop this because everytime i had new ideas.

but
1- ben pridmore use 3 images per loci and he doesnt use it in the same way as everybody just as a back screen.
2- you can improve your loci as time goes by. and you will need a technique simmilar to loci to use in spoken numbers (as you cant review). if you dont have one, you have to use loci. or find a solution.
3- i like using loci as back screen but i discovered that if i have a logical place for the image the loci will work perfectly.
ex: stairs inside the elevator. i just know i will remember that.

i guess you can have 10 loci and make a choice of not using some of them.

or

you can have the same loci as your 100 digits images.

thats what i guess

i wll post something about memorizing names and faces, cause i guess it will make a lot of difference.

1 Like
16

Note that I have no interest in competitions; I just started learning about mnemonics for daily usage, because there’s no better way to learn lists, now lists of +15 items are not difficult to remember even after months.

1.- Like ben, I found only 3 images per loci is best. I suspect the reason ben uses images that way is mostly for speed, since at least I am better with 3D objects than images on walls.

2.- Connecting a key with the stored information is still my main limitation, however with practice the limitation is small enough for my purposes, as if I recall the information for 3-4 days I’m able to learn it. It happens much less frequent now, say in about 1 every 15 items.

3.- I have changed my approach with mnemonics:

  1. 3 items or less, 4 easy obvious items, or items that are a mnemonic in itself (for example, the 5 most frequently found types of hepatitis viruses in descending order are A B C D E) , deserve no mnemonic adaptation.
  2. If each item in the list can be abstracted to a single letter, I make up a word with it (for example, the genes that are associated with the major histocompatibility complex II are HLA-DM, HLA-DO, HLA-DP, HLA-DQ, HLA-DR, so I just remember HLA-D + MODQR, or the genes for the histocompatibility complex I (one) are HLA- + BECK FLAG).
  3. Abstract lists get put in a loci in groups; the prime example of these are numbers, and groups of chemical names, essentially things that you want to learn literally. I do not use pegs, because then you also have to maintain the pegs by recalling all combinations every few days, and the major boost in recall comes from loci, though pegs do improve recall at the cost of a small conversion from the peg to the meaning while remembering. Now that I don't use pegs, I literally imagine 3 numbers floating mid-air per loci. Most often I use ficticious loci, which can be somehow associated with the list, for example the last was about biochemical molecules, so I imagined 3 rooms with 3 items each, where the wall texture is like the molecule models I saw when I was in school; the first room is red, the second blue, and the third green.
  4. Lists where the meaning of the words are the important part, get a visual action sequences, one action per item, where one action is linked somehow to the next. If sounds, touch, smell, taste, or an object seem an obviously good representative of the item I use these, but they are not required. There's also no requirement to stay in the same location, but it naturally happens most of the time. Subjectively, I don't detect a difference in performance between the loci method and visual actions sequences for this kind of list, and the loci method is more expensive. This is also the mnemonic that I end up using most frequently, by far.