3 ms·

I don’t buy it. A huge amount of my understanding of technical topics is visual/graphical (but definitely “conscious”!), and takes significant amounts of effort

I don’t buy it. A huge amount of my understanding of technical topics is visual/graphical (but definitely “conscious”!), and takes significant amounts of effort to serialize and simplify into something that I can write down (or speak) in language or symbols, to the point that it’s often not worth bothering unless I’m trying to explain myself to someone else. If I’m just trying to externalize such topics so I can reason about them with less stress to working memory, vague schematic doodles are often more useful than written prose. Usually explanations of technical solutions to problems (e.g. math proofs or code comments) leave most of the actual understanding unstated, and someone trying to understand the solution for themselves has to painstakingly reconstruct their own understanding, which often ends up somewhat different than the author’s. Many conscious graphical memories and structures (e.g. details of the appearance of someone’s face) are all but impossible to faithfully relate, except sometimes through drawing or the like for expert illustrators.

adrianratnapala 9y ago
> I don’t buy it. A huge amount of my understanding of technical topics is visual/graphical (but definitely “conscious”!), and takes significant ... You're quite right and I'm sorry my comment seened much more "everything is text" than I really wanted. To me the central thing is not those verbal reports, it is the recording into symbolic memory. The symbols need not be verbal, they can be records of our internal representations of geometry or whatever. But whatever they are, they are "parsable" -- i.e. it's not just neural wiring that makes you able to do stuff like catching a ball, it's stuff that you could re-interpret (imprefectly) into a verbal report. And you are right, if the thing was a record of geometry, then you are almost always better off drawing a picture than trying to describe it in words.

adrianratnapala 9y ago

> I don’t buy it. A huge amount of my understanding of technical topics is visual/graphical (but definitely “conscious”!), and takes significant ... You're quite right and I'm sorry my comment seened much more "everything is text" than I really wanted. To me the central thing is not those verbal reports, it is the recording into symbolic memory. The symbols need not be verbal, they can be records of our internal representations of geometry or whatever. But whatever they are, they are "parsable" -- i.e. it's not just neural wiring that makes you able to do stuff like catching a ball, it's stuff that you could re-interpret (imprefectly) into a verbal report. And you are right, if the thing was a record of geometry, then you are almost always better off drawing a picture than trying to describe it in words.