There is a moment in any process of deep learning when linear text stops being enough. You have a collection of ideas, each with its own note, but you feel there is something you cannot see: the way all those ideas relate to each other, the patterns that emerge, the structures that underlie them.
Knowledge maps are a response to that need. They are not a substitute for notes, but a complement that adds a visual and spatial dimension to thinking.
Why visualise knowledge
The human brain processes visual information differently from textual information. A diagram can reveal relationships that linear text obscures. A spatial position can encode information (this is more central, that is more peripheral) that would require many words to express in text.
Visualisation also facilitates creative thinking: when you can see all the elements at once, rather than reading them sequentially, you are more likely to detect unexpected connections.
Mind maps
The mind map is the best-known and oldest format. It starts from a central concept in the middle of the page, from which branches radiate with related ideas, each of which can generate sub-branches.
What it works well for:
- Exploring a topic from a starting point, generating associations.
- Planning a text or presentation: seeing the global structure before writing.
- Brainstorming: capturing ideas without censorship, letting the map grow freely.
Its limitations:
- The radial structure imposes a hierarchy that does not always exist in real knowledge.
- It is difficult to show relationships between branches that do not pass through the central node.
Concept maps
The concept map was developed by Joseph Novak in the 1970s as a learning tool. Unlike the mind map, it has no single centre or radial structure: concepts are connected by labelled arrows that describe the relationship between them.
What it works well for:
- Representing complex knowledge with multiple types of relationships.
- Studying a field and making connections between concepts explicit.
- Identifying gaps in your own knowledge (what I cannot connect, I have not understood).
Its limitations:
- It requires more design effort than the mind map.
- It can become complex and hard to read when the number of nodes grows.
Note graphs
Note graphs are the format most native to modern digital systems. Tools like Obsidian or Roam Research automatically display a graph where each note is a node and each link between notes is an edge.
What it works well for:
- Visualising the emergent structure of an interconnected note system.
- Identifying isolated notes (with no connection to any other) and central notes (highly connected).
- Detecting clusters: groups of highly interconnected notes that suggest dense areas of knowledge.
Its limitations:
- The graph reflects the structure you have built, not the structure of objective knowledge. If you have not created links, the graph will be empty.
- It can become unreadable when the system has thousands of notes.
When to use each format
- Mind map: when you want to explore from a starting point or plan before writing.
- Concept map: when you want to understand a new field or clarify complex relationships.
- Note graph: when you want to see the global structure of your system and find non-obvious connections.
They are not mutually exclusive. You can create a mind map to plan, a concept map to understand, and consult the graph to see where everything fits in your wider system.
The map is not the territory
A final warning: maps are thinking tools, not the thinking itself. A spectacular mind map does not equal deep understanding. A densely connected note graph does not guarantee that the connections are meaningful.
The value of a knowledge map lies in the process of creating it (the thinking that happens while you build it) and in what you discover once you look at it complete. The visual result is only the trace of that thinking.
In the next chapter we go beyond visualisation and explore how to connect ideas from completely different domains: the foundation of real creativity.
