What would be the best techinque to study for this kind of information (details inside) ?


Hey everyone! Hope this is the right place for posting this…

Sorry in advance for the somewhat longish post, please allow me to explain what it is I’m interested in and frame the problem I’ve run into.

I’m an art student working to improve my abilities, and a big obstacle I’ve run into along my journey is memory drawing. I’m pretty comfortable with my artistic skills when I’m looking at a subject in front of me, but take it away - even if it’s a simplish cartoon - and the task of recalling that information becomes much more difficult.

I have devised a way to extract information from my subject, the problem I’m running into is actually remembering it all. What I do is something like this (for a cartoon line drawing at least, which is my goal for the moment):

Break down the object into simple shapes.
Analyze the major lines: list the major angles, break down curved lines into a series of angles to remember
Measure proportions and find the position of elements in relation to one another
Break down the details into small shapes and repeat the process of the above steps until all lines and shapes are accounted for
Finally, memorize the order in which all this information was obtained to keep it step-by-step

Basically, this approach is all about memorizing lines and shapes. The information is visual to a degree, but I’m able to make it logical as well: for instance, knowing that a line rests between 1:00 and 2:00 on a clock can help me recall what angle to place where. I don’t know how photographic memory works, I just know I don’t have one, so that’s why I’ve devised this system of making lots of small logical observations that accumulate.

So I have a system that in theory gives me all the necessary information to recreate the drawing on my own. The problem is it can take up to half an hour to 45 minutes to break down even a simple drawing, and I’m finding that after doing so, I’m only able to recall about 75% of what I saw and noted. So there’s some success, but I would love to remember 100% – so that inconsistencies in my recreated drawing would be failure in the observation stage rather than the memory stage.

I’ve read a little about memory techniques such as the memory palace, but I’m not sure if that would be appropriate for this sort of information since the information all seems to relate to one another.

I was wondering what techniques you folks would recommend I look into in order to retain all of my observations? Any suggestions would definitely be appreciated.

Thanks in advance!



Just writing ideas down; perhaps create a system for describing various types of ‘edges’ of objects, as well as their positioning in the drawing, and their angle and size. Ie: a small rectangular surface at a 1 o’ clock angle in the upper left side of the image.

Would take a lot of work, and practice, so unless you really want to memorize a lot of images in your mind in a way that you can roughly recreate them, I wouldn’t recommend it.



Ok, this art question really started to interest me, because it seems to be a really tough thing (and I myself have always been unskilled drawer + protan).

3 reasons why you can’t draw

3. ...To see something as an artist sees it, you have to look at the Abstract elements within it. Try not to focus solely on the object, try not to label an object but just see it as simple shapes. Abstract elements once drawn then become real in front of your eyes and the left brain will then fire up to make sense of the shapes and label it....
This shape thing seems to be something I've never managed to do. I've always thinked in objects.

I have no idea how an artist sees a subject he is going to draw, but I think that “a complex of major lines and shapes” may be quite right. Although my opinion may be biased because I’m Da Vinci Demons fan, and it constantly depicts how Leonardo sees the world around him in drawings (or in a bunch of lines/shapes).

It actually gave me idea to try the same in the real world - see the important parts of an object through Leonardo’s eyes, as a blueprint. This makes complex things more easily understandable, because in my mind they are separated into small independent pieces. And this is really important in physics/engineering - we have to sketch these little pieces.
Leonardo’s/Engineering Vision is quite similar to Assassin’s Creed Vision (both have bright objects in dark background), only more detailed (divides stuff into really small details).
How Leonardo sees the world (really cool in my opinion)

But using mnemonics to memorize these shapes/lines? Everything already is 100% visual, so there is not much do do about that. Perhaps rapidly dividing the picture into shapes/lines in very organized way (a journey) and simultaneously mentally using your hand to re-create these shapes/lines in your mind. Now the lines are in your visual memory and also doubly in your procedural memory (seeing lines appear in timeline+ handdrawing them).

I can’t think of a way how to memorize all the angles and edges with a system, I think you’d have to robot for that (cause that’s how robots do it, they have zero “right sided brain”, making them opposite to humans. Art is about creativity/right-sided brain, isn’t it?). So I agree with Bateman in this point.


This is necroposting and I’m not sure whether anybody still cares, but it’s a topic I’ve been considering a lot lately.

Doing so in exact details is basically impossible given the sheer volume of “things”, unless, like r30 said, you are a robot, or perhaps a CMOS chip (but then you’re remembering pixels for raster or formulas for vector). I have only been able to come up with two possibilities. The first is sheer bruteforce training. If you look at how Da Vinci describes it (interestingly as does Tesla) in his journals the process is really simple. You look at something, close your eyes, and try and recreate it in your mind in as much detail as possible. To some extent this is the preferred method in my opinion. It works on the principle of feedback loops within the mind. If one considers that the mind tries to build accurate models of reality in order to be able to generalise and make predictions, the one key mechanism required for such a system is some sort of feedback loop. You can’t possibly know whether your model is accurate if you never compare it with reality. Keeping with the theme of the site and taking memorising numbers as an example. It’s really simple to compare, you look at the number, you go through your system for memorising it. You wait a little and then compare what you remember with what’s actually there. If the two don’t match up you change something and try again. If you can work out a system for what to change this will go really quickly. The reason I believe that most people are so poor at remembering what they see (in exact terms) is because unless they’re artists they never practice this feedback loop and given that the brain will never waste energy on something it doesn’t use it doesn’t learn it. This is the method I believe to be the best.

The second method, which I’m still trying to work out, is iconographic so that you can utilise mnemonic “triggers” to get at the memory. This is basically what everybody in this thread has mentioned already. However I think that in order for such a system to be effective it needs to be a lot like building something. So you see two parallel lines 3mm apart 2cm from the top left corner at a 30 degree angle. Rather than trying to remember that (which as the OP mentions, will both take an eternity and details will be missed), you build a larger image from it. I’m not good at describing my thoughts, but I’ll try my best. You could probably draw a decent generic skyscraper from memory alone. All those hundreds of details are fairly easy to do, even adding perspective is not particularly difficult. The reason for this is because you’re not remembering individual lines and grains of concrete, you are remembering a system of relations. A basic description of generic xyz skyscraper is that it’s taller than it is wide, you could probably stack at least ten or twenty (perhaps more) people on top of one another to equal it in size, the sides (if one ignores perspective distortion) are parallel, etc. This is the only way I have managed to find to do something like this. What becomes even more fun is trying to do it with moving systems, like bio-kinetics. Or other senses, like hearing or smell…

I honestly have no idea how to do that. I believe the logical way to approach it would be to build primitives, which combine to form more complex… don’t know what to call it, and eventually creates something that vaguely approximates the image you are seeing. Even less clear to me is how to go through this process quickly. I know it can be done, I can take pretty complex things like computers (and relatively simple software as well) apart in my mind nearly instantly, I can also instantly fiddle with parts and know what the outcomes would be and diagnose various systems. This is all nearly instantaneous. So something in your mind is good at these things, I just don’t know what, and I don’t know how to activate it.


I´m going to try to make something similar, just for sculptures, this is my aproaching:

  1. Make a dictionary of basic faces, forms, etc… linked to the emotion it evocates.
  2. Recognize the most important details of diference between the model you are looking and the picture you had in the dictionary, from the most looking to least.
  3. Recall the basic model, and the important differences with link system, so if you aren´t in a championship of photographic memory it can works fine.

I´m going to try this approach along this year to learn making new models, I accept suggestions.


I don’t have better suggestions than your approach, I rather like the idea of having a basic forms dictionary and then just noting unusual features to add to the base model. Makes the entire approach far simpler and more compact. :slight_smile:

I would be very interested in knowing how you approached making the dictionary. My approach so far as has been very low level, although some of what I’ve been trying to work out is related to mechanics and thus constraints on motion is a fairly important part. Haven’t made much progress so far.


I haven´t made it yet, but by my experience, the best is doing with something similar.
I´ll probably use the options of some personalization avatar from chat, and after adding and substracting some of them just to get one of your like.
There is a law which says that in art you can´t copy exactly what other person did, because it´s an act of expression, so you the most you personalize your dictionary, the better must work.
My idea to make the dictionary is begin with one done which likes to me, and after practising drawing/sculpting it, when some calls you. It´s a good addition, give it a name and link it to your memory palace for this dictionary.
That´s my idea.