Drawing Images from Observation
Updated 2026-08-22
INTRODUCTION
English translation pending.
CORE DEFINITION
From the Great Commentary on the I Ching, which describes the ancient sages observing patterns in heaven, forms on earth, and the near and distant world, and drawing images from them. The core proposition is that understanding advances by abstraction: careful observation yields features that can be represented as symbols, and symbols make an otherwise unbounded reality manipulable. The key qualification is fidelity of abstraction; an image that captures incidental features rather than the governing ones produces a model that is elegant but misleading.
SCAFFOLDING EFFECT
Reduce cognitive load
- Feature extraction: identify which observed properties actually govern the behavior you care about. - Symbol design: compress those properties into a representation you can manipulate and communicate. - Fidelity check: test the symbol against new observations to see whether it still predicts.
Anchor fast decisions
Reality contains more detail than can be held or reasoned about, so understanding requires discarding most of it. Selecting features that govern behavior produces a compact representation, and because the representation is smaller than the thing it stands for, it can be manipulated, compared, and communicated in ways the original cannot. The mechanism carries its own risk: whatever the selection discards becomes invisible in all later reasoning, so an error in the choice of features propagates silently through everything built on the symbol.
MINIMUM ACTION
In progress 0/1Practice this model in one real situation:
account_treeGenealogyexpand_more
menu_bookReferencesexpand_more
Source support: Explicit
- zh.wikipedia.orghttps://zh.wikipedia.org/wiki/%E6%98%93%E7%BB%8Fverified
PRIVATE NOTES · Only visible to you
SAVED Q&A
ENTRY Q&A · Private saving available
Ask with a clear boundary
thinkingmodels answers from published entry context only.
Your question is sent to thinkingmodels. The answer uses public entry context only.
RELATED MODELS