Question on major system

So, I’ve been experimenting with the major system. Most books recommend learning a vocabulary of images for the numbers from 0-100.

My question is, are you supposed to learn only one image per number or more than one image to reduce interference and confusion? The problem is that requires much more work. If you learn 3 images per number, you’ll have to memorize 3 times as many images.

For example, the number 6149234949 translates to sheet-rope-gnome-rope-rope. How are you supposed to imagine this? You would probably do better by sheet-rope-gnome-rib-rob. How do you guys approach this issue?

Thanks in advance.

Hi @rajiv_rewrapped, welcome to the forum!

Typically the way to go is to learn a single Person and a single Action and a single Object that associate with each number. This is referred to as a “PAO list.” If using the major system, (which I heartily endorse!) you can base all of these words on the major phonetics for direct independent learning, which in my opinion is ideal!

So 13’s person could be “TiM.” 74’s action could be “CaRRying.” 49’s object could be “RoPe” (or RiB as you’ve already created!)

Once you have all of your people (or really “characters” to be more exact, since they could be non-human sentient beings, and all your actions, and all your objects learned, (300 elements in total), you can apply the powerful PAO method to learning numbers!

Briefly, a PAO approach involves breaking longer sequences of numbers into chunks that will encode into a scene or scenes that follow a format of “Person does Action to/with Object.”

So your example: 61-49-23-49-49

61 would represent the person (JeD, for example)
49 would be the action (RuBBing)
23 would be the object (gNoMe)

The final 4949 sequence would start the PAO sequence again:

49 person (RoB)
49 action (RuBBing)

This format would generate the following scene that is incredibly easy to visualize and remember:

“JeD is RuBBing a gNoMe and RoB is RuBBing himself.”

(Rob performs the action on himself because there is no “object” number in the series. If there were two more digits, then the second part of the scene would be in a complete PAO sequence. So, if you had 614923494930, you could arrange it as JeD is RuBBing a gNoMe and RoB is RuBBing a MooSe.)

How you visualize that is up to you, whether you take it as two separate scenes or combine them into a single longer event, but this is a powerful way to memorize long strings of numbers nearly effortlessly. The system provides the structure and automatically generates absurd memorable images.

Think of it like a gameshow that has three big spinner wheels each containing 100 spaces for each number 00-99. The first is the “people” wheel, the 2-digit number that it lands on gives you the person for your scene, then the second wheel spins and gices you the action that the person is doing, and then the final wheel spins and gives you the object that is affected by the action.

It does require more work, but thats not a problem, it’s the cost of entry to be able to use the system. All memory systems require effort and you get out what you put in.

Don’t be put off by the requirement of learning 300 independent associations in order to acquire this system. The benefits are sooo worth the effort. 300 is really not all that many associations to make if you think about it… How many phone numbers, addresses, email addresses, websites, dates in history, birthdays, sports team stats, etc have you committed to memory? All of these are abstract information that you’ve learned to associate with imagery. I’m willing to bet there are hundreds and hundreds of those associations already living in your brain. Whats a couple hundred more? Especially when there is a step by step system for creating and learning them ready made for you.

Take them in chunks of 12 per day, thats 4 people, 4 actions, 4 objects. Learn new ones each day for 5 days, then spend 2 days reviewing the full group. Within 5 weeks, you’ll have seen them all and be pretty familiar with them. Within 2-3 months of consistent review and practice you’ll be pretty fluent. This is a super conservative time frame, you can likely go quicker than this if you want.

Be realistic about the timeframe and you won’t be overwhelmed or discouraged at what seems to be “slow” progress. You’ll improve incrementally and before you know it you’ll have an awesome tool at your disposal.

This, like all memory techniques are like shortcuts. But you have to learn the shortcut first before you can use it effectively. There is no shortcut to learning these shortcuts. They’ll take a good amount of time and effort up front before they are ready, but once you’ve learned them, you’ll reap the benefits many times over when you use them.

Note: there has been discussion about whether to structure your scenes in Person-Action-Object format or Person-Object-Action format. POA scenes could be thought of as “I have a Person and an Object. They interact via an Action.”

Some like one way better and some like the other. Some of it depends on the way the grammar of your native language is structured, some consider the way that scenes are progressively built in the mind when ordering the action and object. There is no “correct” or “better” way when it comes to individual ewse of use, it all comes down to trying and seeing what clicks for you the best.

4 Likes

@TheHumanTim
I really appreciate your efforts at writing this thorough explanation PAO. However, I already knew how PAO works. You’re suggesting I upgrade the two digit system to a six digit system.

The point is that in my life, I rarely have to remember phone numbers or other long string of numbers. What I need to remember are history years and dates. So, I thought I could get by with a two digit system.

Old memory books like those of Harry Lorayne propose the two digit major system as a complete one that could be stretched to remember phone numbers etc. But Lorayne never tackles this issue squarely - and this is a doubt I had from the very beginning - i.e. what happens if you use the same images or the same pegs over and over, or use them in close succession. I’ve also considered representing things with a ‘category’ rather than a single object in the hope it will allow for more variety in the combined associative images. But then I have less material available for representing other things.

However, here’s my question to you: when you’re remembering some numbers with PAO, and the numbers are 211, 212, 213, (let’s the numbers represent 2nd November, 2nd December etc.) don’t you think the images you create would somewhat interfere with one another? If yes, what strategies do you use to reduce confusion?

No, this is not a 6-digit system. It is a 3-element 2-digit system. Yes, there is a big difference. The number of digits in a system refers to how many digits are encoded into a single element. Elements are combined into images and scenes.

More here:

Not at all. In this case I’d use my 3-digit system: aNTeaTer, NaTHaN, and NaughTyMeg. Or, if using a 1 or 2 digit PAO system, you can make person 2 interact with objects 11, 12 and 13, or person 21 interact with object 1, 2, and 3. This would be quite easy to recall properly. If you need to store each as independent items, using 3 consecutive loci makes it easier still.

For 4-digit years, you could designate a person for each century and an object for each decade. So 1925 is person 19 object 25. 615 is person 06 object 15. Tons of ways to slice this, but using a PAO list makes it simple however you want to tackle it.

Again, if you’ve already learned 100 things that you associate to 00-99, just learn the other 200 to fill up the full list. Its really not that much more at all. It will give you tons of non-repetitive combinations and variety to bring to your imagery. If not, you’ll be stuck with repetitive imagery and constantly be dealing with the problem of sameness.

2 Likes

He also says really nothing about arguably the most powerful general memory technique, the memory palace or method of loci. Lorayne was a big proponent of story linking, and somewhat “on-the-fly” creation of words/images when it came to numbers, which is serviceable but in my opinion is inferior to a PAO structure for the same reasons you’ve identified. Repetition, problem of sameness, and extended sequences of links being challenging. If you lose a link you run the risk of losing everything that comes after. His instruction on number systems was pretty basic, the techniques have advanced quite a bit since The Memory Book was published.

@rajiv_rewrapped

There are three major ways to visualize sequences of numbers:

  • Merging a group in one image (a ghost/sheet tries to lasso/rope a gnome)
  • Linking multiple images in a story (a ghost unties a rope holding a door in a haunted house. The rope pulls a gnome down from the fireplace. The gnome gets his foot caught in another rope and grabs the rope cord from the fireplace which slides open the fireplace.)
  • Associating with a peg (a sheet covers my head, a rope hangs out of my nose, a gnome walks around my teeth in my mouth, a rope binds around my heart, a rope is a belt around my loins)

The duplicate number problem occurs only in the first two where the images would not be too unique. In a peg system the association with a unique peg makes the whole image fairly unique. That’s why @TheHumanTim is favoring the PAO although you could simplify it some by just using unique images for a 2-digit or 3-digit system.

Doug

With PAO how do you link the first image to the second and so on for long numbers? In your example of gnome rubbing you just said “and” to connect the two images.

You typically make 6-digit chunks with a PAO. Link the chunks however you want. Direct link or loci method are the two easiest.

61-49-23-49-49

Link:
Jed rubs the gnome, the gnome interacts with Rob as he rubs himself. Maybe the gnome is freaked out by Jed rubbing it and tries to get away but stumbles into Rob.

Loci:
Front door is loci 1: Jed crashes through the front door while rubbing the gnome.

Entryway closet is loci 2: Inside the closet, Rob has a coat hanger through his shirt and is hanging on the closet rod. He’s startled since he wasn’t expecting anyone to find him in there rubbing himself.

Both create memorable and vivid connected scenes.

So, have you created a PAO system AND a 3 digit system separately? That means learning ~900 images separately.

So to clarify (with linking), after you have “Jed rubs the gnome” the next image is the gnome interacting with Rob, like bumping into him and THEN we see the image of Rob rubbing himself, no? Otherwise, it seems like you’d have to remember that Rob is rubbing himself before you know to have the gnome interact with him. With loci I can see that the front door need not have any connection to the closet in order for the recall to work.

Yes.

But, actually its an expanded LPAAO (location, person, action, adjective, object) system for 0-9 and 00-99, so 50 words/images total for the 1-digit system and 500 for the 2-digit system. For my 3-digit system it’s just one word per number, so 1000 for that. That makes 1550 total unique independent associations for numbers. All based on major system phonetics so they are pretty simple to sound out, read, and learn.

I’ve posted my complete 1 and 2 digit word lists here if interested:

At some point I may try for a 3-digit Person/Object list, which would be an additional 1000.

It may seem like a lot, but I didn’t come up with them or learn them fluently in one week. It took several months of effort before they were usable. I was realistic about my goals and expectations and took one section at a time. Now that they’ve been learned, its no problem to maintain and use them. Like I’ve said, the initial effort is worth the end result.

Yeah. This is simply because after 6 digits, the PAO sequence resets, so if you’re not using loci for each 6-digit chunk, you need a way to recall what the starting point of the next chunk is. You know it should be a person, but what is triggering your recall of who that person is? Without some way to get from the end of one set to the beginning of the other, it fails.

if you direct link PAO-PAO, you almost need to make it PAO-A-PAO to make the connection. I suppose thats an idea, so instead of jed rub gnome rob rub, you could just do JeD RuB gNoMe RuB RoB. Then the action that links the two actually encodes info in addition to providing the link and isn’t just a filler connector.

As you said, with loci, you don’t really care about direct connections between the elements in the sets of 6, the flow of the loci provide those links. I prefer the loci method since the sequence of loci are strong and set ahead of time. I know that door flows to closet flows to couch flows to tv, etc. I don’t have to spend any energy trying to come up with links between the loci, they’re just there from my familiarity with the spaces. With direct linking (non loci), there is also sometimes less of a context surrounding the story to help bring it to life or at least its harder to generate one on the fly especially if trying to go really fast.

1 Like

Not to hijack this thread, but I have to say the loci concept is new to me. I’ve been reading up on it and I don’t understand how it is possible to have a palace or palaces with easily remembered loci for, say, 500 items like pi. Is there one detailed resource that goes into it in detail, or is it just a matter of reading up on the posts in this forum?

You could also have images for every month. They’re just 12.

Think about how many physical places you are familiar with. Think of all of the places you’ve lived, all of your relatives and friends and their houses. Your job locations, your schools, a doctor’s office. A supermarket or store. A park or playground that you’ve frequented. Airports, stadiums, concert venues, bowling alleys… literally any place you can think of.

If you can picture yourself navigating through those places and can think of distinct landmarks or “loci” within each, you’ve got yourself some memory palaces! (Landmarks or “loci” can be large or small, entire rooms, or smaller areas or objects within them.)

You can even use fictitious or virtual locations for your palaces or places you’ve never even been to. These can come in many forms, from movie/book or game locations, virtual tours of famous places online, or just spaces that you can make up entirely from your own imagination. If you can picture yourself there and figure out a logical path through the area with unique stopping points along the way, you can use it as a memory palace.

You can use palaces individually or come up with a sensible way to connect a journey between many palaces to form a giant meta-palace or “memory city” or whatever you’d like to call it with hundreds or even thousands of loci that are organized and structured by the path you take through them.

When it comes to using them, if you can have the representative imagery that you’ve encoded interact with each loci in some way so that as you think of the loci, it will trigger the recall of the images. I like to use the example of a front door to illustrate this. If the image you’re trying to remember is of a tiger, imagine the tiger smashing through the door, shards of wood flying everywhere, pelting you in the face. This way, you imagine the door, then you hopefully recall that something smashed through the door, and that should lead you to the tiger. If, on the other hand, you just imagine opening the door and there’s a tiger sitting there, you’ve got nothing to really connect that door to the tiger and it becomes much tougher to recall the connection.

To encode 500 digits using a standard PAO for 6 digits per scene, you’d only need 84 loci. If each of your palace areas contains just 10 loci, you’d only need 9 areas to get to this number. With something like Pi, you probably aren’t going for speed so you can take your time crafting the loci and the interactions. You can document the order of loci along with a quick description of each in an excel doc or notebook to help review initially. You’d be surprised how easy it can be.

When I memorized Pi, I stored over 4000 sequential digits across more than 500 loci. These loci were grouped into about 60 general areas. I made scenes following a “person action adjective object” format, encoding two digits for each element, so 8 digits were stored at each loci. These scenes were tied to the loci with vivid interactions which allowed me to easily differentiate what was happening at each and avoided issues with sameness and repetition. This works because having the scenes occur in different areas provides a unique context to each of them and allows for the same elements to occur in multiple instances in different spaces and times within the sequence. If you’re interested you can check out the list of my loci for pi below to see how something of this size can be laid out.

Pi Loci

OUTSIDE STADIUM
road into met life stadium
parking attendant booth
stadium parking space
tailgate camper
tailgate tent
tailgate grill
tailgate cornhole
security bag check
metal detectors
stadium ticket scanner

STADIUM GROUNDS
designated driver signup
cover band stage
dive and catch
radio stage
target throw
sausage stand
main arch
restroom door
tunnel
fancy club window

GIFT SHOP
stadium gift shop door
jerseys
sweatshirts
knicknacks
posters
bottles glasses koozies
outdoor decor
autographed helmets
checkout line
checkout counter

STADIUM CONCOURSE
escalator up
captain morgan statue
museum wait line
museum door
superbowl trophies
cleats
vintage programs
vintage helmets
tv showing classic games
old giants stadium seat

CLUB ZONE
club door
club bathroom doors
free snack basket
drink counter
lounge
latke station
soup/sliders
condiment cart
sushi guy
hightop tables

SEATING SECTION
section door
stairs down
seats
cup holder
railing
g-line to the left
jumbotron left
broadcast booth
hot dog guy in stands
right jumbotron

ON FIELD
tunnel / smoke machine
fieldgoal net
near goalpost
near endzone
down marker banner
home bench
med tent
kicking net
gatoraid table
mid field

AMUSEMENT PARK ENTRANCE
park gate
compounce parking space
sidewalk bench on way up
tunnel
badge scanner
compounce ticket giver
chalk drawing
water bottle table
wristband checkin
turnstile

MIDWAY
chicken tender stand
skiball
water racer
bucket toss
stroller rental
tattoo/facepaint
characature
rubiks cube guy
benches
corner gift shop

WATER PARK ENTRANCE
locker rental
lockers
rope fence by changing
changing area entrance
changing rooms
changing bathroom urinal
changing bathroom stall
changing bathroom sink

WATER RIDES
wavepool
kid wading pool
little kid water slide
triple racer kids slide
lighthouse thrill slide

BEACH AND KID ZONE
beach volleyball
beach jenga
lake swimming
beach chair area
beer cart
kid pirate ship stuff
dumper bucket

FOOD ZONE
pizza building by water
front patio
statues
booth
bar
bathroom door
soda machine
back door
back patio tables
band gazebo
lazy river

PATH TO COOKOUT ZONE
dippin dots
basketball game
walkway up to dinner
golf into lake

COOKOUT AREA
dinner ticket taker
dinner entrance
dinner buffet
dinner cookies
hot dog stand
photobooth
picnic tables
giant cornhole
dj stand

PATHWAY BACK TO PARK
floom ride
train down
train station
carousel
bumper cars
funnel cake

MAIN PARK RIDES
drop tower
wooden coaster
ferris wheel
viking swinging ship
dinosaurs
kiddie coaster
flying elephants
magic show
soda machine
hand stamper exit

GROCERY STORE LOT
traffic light into store
stop sign
parking spot grassy divider
cart return
ev charger
propane tanks
salvation army guy

STORE ENTRANCE
auto door in
bottle return machines
bottle return garbage
scanner
florist checkout
flower racks
customer service
pharmacy
bakery cake display
bakery counter

FRONT WALL AISLE
roll case
garlic bread
cheese
dips
yogurt
bread
milk

FRIDGE AISLE 1
lemonaide
butter
eggs
frozen sausage
chicken
pork

BACK WALL AISLE
ground beef
steaks
butcher
fish sticks
seafood lobsters
hotdogs
bacon

DELI AND PRODUCE
deli
rotisserie chicken
prepared foods
sushi
lettuce
cucumbers
mushrooms
ice
cashier checkout
exit door

STORE PLAZA WITH SUSHI PLACE
sidewalk
restauraint specials sign
restauraunt entrance
entrance fridge
cashier
tv
sushi chef
sushi bar seat
back kitchen
table

PATH TO THE PARK
stop sign side ext
sidewalk on right to park
tennis court
basketball court
park entrance
gazebo

OUTBUILDING AND PARKING
bathroom building
sink
urinal
stall
guard chair
guard hut
speedbump
handicapped sign
parking space
parking railing

PATH TO THE LAKE
hill
down path
flood barrier mound
tetherball
camp grill
lifeguard shed
playground swing
playground steps up
climbing wall

BEACH AND WATER
picnic table
sandbag line
beach towel spot
sand castle
shallow water rocks
standing in water
to right ropes
inflatable swan
far line buoy
lifeguard chair

LODGE BUILDING
trash cans
stairs up to beach lodge
main floor in lodge
first aid closet
bathrooms
water cooler
door out to deck
rail on deck

CAMP WATER AREA
camp swim zone
jetty
boulders
boat launch ramp
boat rules sign
kayak rack

PAVILLION AREA
elks sign
elks pavillion
elks grill
elks table
back exit

OUTSIDE LIBRARY
library parking entrance
up hill parking space
up hill bench
shed
generator
side sidewalk
outside book drop
outside table
take and make pickup

LIBRARY ENTRANCE
button for door
auto door in
airlock

LIBRARY ROOM
community room
stage step
flags
projector screen
piano
photo monitor
photog table
back door
alley
camera
photog floor
greenscreen backdrop

HALLWAY TO MAIN LIBRARY AREA
book sale rack
water fountain
corkboard
bathrooms
art display case

RECEPTION AND SECTIONS
book drop
checkout line
prize rollaround cart
young adult section
religious books
coin encyclopedia
card magic book
magazines
fantasy rotating display
sci fi
mystery

RESOURCE AND KID ZONE
prize shelf
computers
copy machine
kids 1000 books board
kid survey board
new books
alphabet wall
kids cookbooks
biographies
scout corner
kids computers
craft table

KID ROOM 2
catalog computer
kids raffle prize shelf
kids board game wall
kids coloring table
backpack checkout
kids picture book
dehumidifier

LITTLE KID ROOM
play kitchen
play area foam floor mats
play area seats
puppet stage
craft cart
sink
cabinets

KID ROOM 3 / CHECKOUT
funhouse mirror
holiday books
lounge chairs
kids checkout
prize bin
kids bathroom
dvds
exit door

TO SENIOR CENTER
senior center door
senior center sidewalk
new house construction
bus stop
bridge
gas station

ART CENTER TAG SALE
arts center sign
parking grass hill
golf clubs
lawn spreader
bikes
skis
kids play table
outside cashier

TAG SALE ROOM 1
halloween lanterns
costume rack
dining tables
hutches
rocking chairs

TAG SALE ROOM 2
glass nicknacks pelican
board games
stuffed animals
puzzles
books
crock pots
dvds

TAG SALE ROOM 3
baskets
kids clothes
ornaments
cashier

LUMBERYARD
lumberyard uphill drive
parking spot
piles of lumber
ramp
door
order guy
sealers and stains
cashier

TRACTOR SHOP
upper driveway
house door
empty house
lower parking spot
gated back entrance
tractors
drop off shed
outside signs
door
counter
cashier
stock room

ROAD TO SCHOOL
stop lights by gas station
fake turkey on right
daycare parking
daycare entrance door
right house dog yard

ELEMENTARY SCHOOL ENTRY
school entrance sign
congrats signs
front field
right front fence writing
baseball bleachers
parking spot
garage
dirt pile
dumpsters

PLAYGROUND
playground entrance
spinner
kiddie slide
play tunnel
whale spring rider
turtle balance
car front
baby swing
ramp
monkey bars
parallel bars
climbing trees
climbing net
ladder
bridge
tube slide
parent bench
sign

UP TO PRESCHOOL ENTRANCE
brick sideawalk
preschool barriers
veggie garden
pickup line
pickup table
pickup door
cafe passthrough
kitchen

CAFE EVENT SETUP
cafe door
lunch table
dj
fan
bleacher
stairs to stage
game host
game contestant

RECEPTION SPACE
luau poster
lost and found
library door
clay sculptures
reading amphitheater
big rocking chair
bathrooms
water fountain
dinosaur bones

OFFICE
office secretary
nurse desk
nurse cot
lobby entrance
tv screen
homework table
tiny door

OUTSIDE FIELD DAY
painted rock pile
balance beam
water relay
limbo
kid basketball
lacrosse
obstacles
t-ball
frisbee

TO THE GYM
gym door
gym mats
roller pads
cargo net
ball rack
parachute
scoreboard

CLASSROOM HALLWAY
projects on the wall
outside class chair
class bay window
class door
play kitchen
book rack
blackboard
kid sink
kid bathroom
math cubes
teacher desk
reading rug
projector
cubby arch
cubby

MUSIC ROOM
music room door
piano
guitar
drums
recorder
tambourine

ART ROOM
art room door
corner by smocks
paints
eisel
woodcarving
clay sculptures
kiln
exit door

PARK TRAIL AREA
park entrance crosswalk
bog
potholes
boulder
map
bench
ridge
goose pond
sheep shed
left ridge path
bike repair
picnic table
right trail
left rushing stream
house yard below
right dripping rocks
right tree root
aquifer pipe
parking zone

TOWN DUMP
dump entrance
guard booth
hopper dumpsters
tv house
air conditioners
scrap metal
brush pile

CEMETARY
groundskeeper shed
cemetary gate
headstone
tomb
open grave
coffin
wreath
audience line
hearse

MINI SHOPPING PLAZA
entrance
pharmacy drivethru lane
pharmacist window
ace propane tank
dunkin drivethrough
ace parking space
leaf blowers
wheelbarrows
grills
ace door
mole traps
birdseed

Wow that’s quite a list! I can see the power of combining a PAO type system with loci. You’re saying this works better than linking or using pegs, even with the additional work needed to set up the palaces. Can you reuse the same palace for more than one long term memory list? I understand about making the same locus different with a new season or raining or whatever. Otherwise, it would seem like a waste if you can really only use a palace once.

How often do you have to go over your pi list to retain it? It would seem harder to retain because you don’t actually use the 4000 list for anything.

Its usually not a “one or the other” approach, its a combination of multiple techniques. The PAO structure (or PAAO or whatever is used) IS linking. Loci ARE pegs in a sense, as having scenes interact with the Loci links them in the same way as a general peg list. The path through the palace determines the pegs, creates links between them, and provides an organizational structure that takes advantage of our innate strength of spacial memory and navigation. By “outsourcing” the job of linking scenes together to that spacial system, it frees up focus and mental effort that can be applied to creatively construct and visualize the representational imagery used to encode our information.

Lots of varying opinions on this. You can try but I usually wouldn’t unless the imagery thats being stored is completely different with no chance of overlapping images. For instance, I memorized some stats about the planets in the solar system, their sizes, gravity, distance from the sun, etc. Lots of number stuff. I created a separate set of palaces for this info, because the imagery would tend to mirror and clash with some of the pi number imagery if I reused the Pi locations.

When I was first starting out I tried memorizing some info about world countries in two ways, and arranged both alphabetically and by population. I tried using the same palace that I’d stored info about the periodic table of the elements. I found that I’d get overlaps where I’d see my country image but then would have this weird false interaction with some of the element stuff that was hanging around the same loci. So I stopped being lazy and just made a new palace and it cleared up both data sets much better.

For short-term stuff like competitive event practice for cards, words, numbers etc, its possible to reuse palaces, but its good to give them a day off in between uses to let then reset and clear out. Some folks use the same palaces for different disciplines as long as there are no shared elements. I wouldn’t use the same palace for numbers and for cards because they use the same imagery for me, but I could use a palace for both numbers and names/faces or cards and words, since there is much less chance of overlap. But I think its cleaner to just designate palaces for a single type of use.

Not at all. I think this is a really common misconception that people have when they’re first starting to use mem palaces. It seems difficult to come up with palaces and loci at first so they want to just hang on to the few they’re comfortable with,Then they try to cram a bunch of data sets into the same loci, they experience interference, and then say the memory palace technique is ineffective or limiting. But really, there are almost limitless amounts of palaces and loci that can be found or created. Literally any location you can think of can be turned into a palace.

Prior to memorizing, you should be organizing your data into a sensible structure for long term storage anyways, so as part of that organizing, determine what your palace and loci structure will be. It can be tough at first, sure, but no tougher than trying to learn a standard peg list or substitution system, and after you make a few palaces and experiement with them, it gets easier and faster, just like any other skill.

I don’t think it’d necessarily be harder to retain since its made up of image/scenes at loci just like everything else. It would just be a matter of smart spaced review scheduling.

Pi was a short term personal challenge project for me to see if I could do it, and it also was a great training exercise to help me learn my number system and visualization techniques. I worked on it for a couple months last summer and built it up progressively, a hundred or so digits per day. I had a goal total in mind and once I hit that I basically stopped reviewing it and moved on to other things.

I actually tried running it a couple weeks ago just for fun and got through the first about 200 digits without a hitch, then it got a little fuzzy shortly after that. This was after probably 10 months of not practicing it at all, so thats about as good as can be expected. If I was serious about maintaining it, I’d plug it into a spaced repetition schedule and I’d be pretty confident it would stick around accurately for quite a while in there depending on how many reps I gave it.

Thanks for the informative reply… lots of good stuff. If I want to create a new palace from an unfamiliar place how well do I need to learn that place? For instance, if I think back on my dorm room from decades ago I can probably come up with about 10 loci. Kind of amazing now that I think about it. However, if I go to a Walmart or a gardening center I’ve never been to is it sufficient to do one visit and then you know it forever?

So there is almost nothing that you can just take a single pass at and “know it forever.” Memory palaces included. They take thought and review and revision just like learning other types of data. The plus side though, is that we are pretty hardwired to be very good at internalizing spaces and if you are deliberate with your attention as you learn the space, it should be very easy to get it to stick. Then of couse if you use and review it it will stick with you.

As for the question of how well you should know a place in order to use it as a palace… Most palaces for me end up as sort of idealized and optimized versions rather than completely accurate one-to-one reproductions.

Walmart is a great example because there are so many areas with unique loci potential. You could take 10 departments (electronics, mens clothes, women’s clothes, toys, automotive, sports, crafting, garden center, food, hardware, etc…) and note 10 or so unique items in each and have yourself a 100 loci journey.

But the cool thing is to be useful as a memory palace it doesn’t have to be completely true to life. You can take things that COULD be found in those sections and just mentally place them in as loci. You can alter the arrangement and spacing of your loci as you see fit to make it flow more intuitively for you. Its kind of like a movie thats “based on true events.” There is creative liberty and editing that is done to polish and strengthen the story, and you can do the same with memory palaces and their loci.

If you have time when you’re in a new space, open up a note app or google doc or whatever and as you physically traverse the space, note the order which you encounter general areas, and then within each, find unique loci that stand out. Note as many as you can observe, you can always edit it down later or adjust placement. Some places like big department stores will provide a ton. Hundreds. A barber shop may only give you 6 or 7, but hey, they’re still a unique list of loci pegs you can use on their own or if you link several broader areas together almost like in a “day on the town” journey, you can quickly build up to dozens and dozens of loci. 7 at the barber shop, 7 at the Burger King, 7 at the dentist… There’s 21 for a solid easy palace.

If you noticed in my Pi loci list, I never really had more than a dozen or so individual loci in each area, I just connected all of those smaller zones together to get to the 500+ total I needed.

Hope this gives you some ideas!

1 Like

How about shooting a walk through video with a phone? That would give me something to look back on and could also use to make a list for review.

So where I’m at is I’m fluent with major system phoenetics (1=s,z 2=n, etc.) and I am solid with 100 pegs. I haven’t used the system enough to know the 100 pegs on sight, or instantly, so I tried something a little different that seems to work. I use the major consonant sounds for each number but use an “a” vowel for every peg. For the first 25 or so I use more standard objects because I already learned them pretty well (like 15 is towel, 22 is nun). However, I am using unusual images like 73=cam or 84=far because I can recall these lesser used images much more easily because I know the vowel is an “ahh” sound.

I’m willing to revamp all of the above if the end result pays good dividends. It sounds like your ideal system is PAAO with loci, right? What would you recommend if I were to start over, still using major? I already learned 525 backgammon 2nd moves so I’ll keep that as is. You are starting to convince me that the method of loci is also the way to go.