Ideal ratio of people to objects (mnemonic images)?

I’ve recently completed a list of 2704 mnemonic images using the ben system, which I’ve been practicing over the past several weeks. The list was originally designed to be exactly 50% objects and 50% “people”, and by people, I mean anything animate (people, animals, etc.). The premise of course was that this 50/50 ratio of people to objects promoted a natural interaction of images. While practicing this system has been going extremely well, there were two things that I didn’t expect. First, rigidly classifying images as either “people” or objects based on whether they’re animate or inanimate isn’t valid for all images. I found that there were many animate images that could both give and receive an action. A dog, for instance, could easily perform an action, but it could also easily receive an action by being pet. The other thing that I found is that humanoid-humanoid linking was more difficult (to form and recall) than object-object linking, which suggests that the ideal ratio of people to objects would actually be < 1.

While most of the list won’t change at this point, I was thinking about making a few changes based on what I’ve found, which leads me to the point of this thread. I’m looking for any input regarding the ideal ratio of people to objects from others with experience.


I think it depends on the person. When I first started creating systems, I tried to make them perfectly-structured, but I found that it was nearly impossible, especially with large systems where you need to find a lot of images. :slight_smile:

1 Like

Just a status report:

So I’ve been working on two things. First, I’ve been trying to see if there’s any difference in my ability to remember people vs objects. As far as I can tell, there isn’t. Second, in order to better promote interaction between images, I’ve been thinking of different ways inanimate objects can take on animate qualities. For instance, a pair of shoes might walk by itself, a puppet could get up and walk, etc. So far it’s been going really well.

Just for fun, here’s the story for the deck I memorized a few minutes ago: The phoenix from Harry Potter lands on a girls back. She then walks forward but immediately notices a large collection of ants which begin crawling up a kangaroo. Startled, the kangaroo hops away but is immediately confronted by an enemy from Yoshi’s island. A kick to the face ends the confrontation, and the enemy then takes off on the Mayflower sailing through a horde off zombies, one of which picks up a high-pressure washer and uses it to scare away a nearby Pokemon, which makes its escape in a steam engine. The steam engine then scares away a baby cougar, which decides to play with a plastic pale. Droopy, noticing the cougar, comments on the situation in his normal monotone manner, but then he turns around and immediately is accosted by a giant gelatinous creature. This creature then picks up a POW block (item from the original Super Mario) and launches it across the room, destroying all the marshmallow candy in its path. Emerging from the candy is a puppet, which walks across the room right into Don (a driver from a video game), who is distracted by the rubies scattered along the floor. There are so many of them, that they nearly trip a nearby ballerina who’s trying to dance. And while the ballerina manages to avoid the rubies, she unfortunately steps on a tack, which she pulls out and tosses at Roger (a ghost from a video game). Roger, now flustered, picks up some iron knuckles and dukes it out with a three-headed snake, who then tips over a hanging bucket (an item from a video game) which falls and lands on a generator, bringing it to life vuhm vuhm vuhm (or whatever sound generators make).

I’ve never actually read somebody else’s story, so I have no idea how normal/abnormal this is.

1 Like

Looks good to me. :slight_smile:

If you haven’t seen it yet, you might be interested in this music video:


Thank you so much for the link! I’m surprised I haven’t seen anything like that before. The way my images interact is nearly identical, and I actually have most of the images that appear in the music video. That’s really cool!

Just an update:

All right, so I’ve memorized a lot of decks at this point, and I’ve found that there are too many inanimate objects in my current system. It’s not way off what I would consider to be the ideal ratio, but it is off. I’m currently working on version 2.0, which will correct this issue. I’m also eliminating all images that I’ve found to be cumbersome or difficult to link. The changes won’t be too extensive, because I think I’m already pretty close to what I’m aiming for, but they will correct any issues I’ve found.

Also, just a few more weeks in school. Once finals are done I plan to really hone my system, hopefully getting some decent times posted. We’ll see.



All right, so I’m just about done putting the final touches on version 2.0. I kept track of all the conversions on a piece of paper. The goal wasn’t to just produce more animate images, but to do so symmetrically. For instance, for every one image converted in the decimal subset I would convert roughly 2.7 images in the card pair set, this way the ratio would be constant among subsets.

Practice runs are going perfectly so far. The ratio for any 26 images is obviously high variance though, so I’m interested in a long-term average, which can only be tested by practice runs.

1 Like