Independent Thinking in the Age of AI — Visual PKM and 4D Knowledge

The biggest risk AI poses isn’t job loss — it’s losing our ability to think independently. Zsolt Viczian’s PKM Summit 2026 talk argues that text-based thinking is like “looking through a keyhole,” while our brains evolved for spatial, visual navigation. His solution: 4D Visual PKM, where notes are visual-first with text as metadata, not the other way around. The core insight — different representations of the same problem lead to fundamentally different solutions — has profound implications for how we use AI without surrendering our cognitive agency.

The Core Thesis: AI Is Doing Our Thinking

Viczian opens with a provocation: the PKM community represents the top 0.1-0.5% of deliberate thinkers globally — and even they are at risk.

The GPS analogy: 20 years of GPS has degraded our spatial navigation skills. Most young people can’t read a paper atlas. The same is happening with thinking:

“The quality of your life is directly proportional to the quality of thinking you do every day.”

The danger isn’t that AI is wrong — it’s that AI is convenient enough that we stop thinking at all. Steve Jobs warned against “living with the results of other people’s thinking.” Today, add: machines’ thinking.

Why Text Is Failing Us

Text is sequential — you absorb it one word at a time while your mind works hard to stitch pieces together. Viczian demonstrates this with a powerful experiment:

The “odd one out” test:

• Given 100 words in a text list, finding the one that doesn’t belong takes significant scanning time
• Given the same items as images (animals + one box), identification is instantaneous
• AI can solve the text version faster than us — but we solve the visual version faster than AI

Text processing:
  word → word → word → word → ... → mental model (slow, sequential)

Visual processing:
  [entire scene] → instant pattern recognition → understanding (parallel)

Why Our Brains Prefer Visuals

“A mind on the hoof” (Annie Murphy Paul’s term): Our ancestors didn’t survive the savannah by reading bullet points. They navigated spatially, recognized patterns visually, and simulated movement mentally.

• 30%+ of your brain is dedicated to visual processing
• We have mirror neurons that simulate observed movement
• We think in space (“the future is ahead of us, the past is behind us”)
• We remember concrete nouns (ball, book, bottle) easier than abstract ones (freedom, peace, ethics)
• Illustrations turn the abstract into the concrete — that’s why they help us think

The Game of 15: Proof That Representation Matters

This is the most compelling part of the talk. Same game, two representations:

Version 1 (text/numbers): Players take turns picking numbers 1-9. First to collect three numbers summing to 15 wins.

Try it — your mind literally burns juggling combinations. Even educated adults struggle.

Version 2 (visual/spatial): Same game, but laid out as a 3×3 magic square (every row, column, diagonal sums to 15). Players mark circles and crosses.

Magic Square:
┌───┬───┬───┐
│ 2 │ 7 │ 6 │
├───┼───┼───┤
│ 9 │ 5 │ 1 │
├───┼───┼───┤
│ 4 │ 3 │ 8 │
└───┴───┴───┘

Same game → now it's just Tic-Tac-Toe!
A 3-year-old can play what adults struggled with.

“Solving a problem simply means representing it so as to make the solution transparent.” — Saul Amarel

Three principles emerge:

1. Different problems benefit from different representations
2. The representation drives the solutions you find
3. Discovery emerges through iterative experimentation with representations

Outsmarting the Lazy Brain (System 1 vs. System 2)

Daniel Kahneman’s framework, applied to knowledge work:

System 1 (fast, lazy):
  Situation → Immediate judgment (pattern match → done)

  Problem: WYSIATI — "What You See Is All There Is"
  Your System 1 convinces System 2 there's nothing more to look for.

System 2 (slow, deliberate):
  Situation → Explore → Collect data points → Then judge

  Problem: Short-term memory holds only 7±2 items
  Solution: Write it down — but text is a "cassette tape" (sequential)
  Better solution: Diagrams = "visual prosthetics for better thinking"

Edward de Bono’s recommendation: inject exploration between seeing and judging. But exploration requires holding many data points simultaneously — which is exactly what visuals excel at.

The cassette tape vs. iTunes analogy:

• Text = cassette tape (rewind, fast forward, linear scanning)
• Visuals = iTunes (see your entire library at a glance, jump to anything instantly)

The Packing Paradox: Why We Organize Bags Better Than Thoughts

When packing for a trip, you:

1. Lay clothes out on the bed
2. Lay toiletries on the counter
3. Inspect everything visually
4. Pack neatly

But when organizing concepts, we immediately switch from “iTunes” (spatial layout) back to “Walkman” (linear text). Why?

Nature vs. nurture:

“The single best way to become a better thinker is by building a library of visual primitives.”

Visual primitives aren’t stick figures — they’re spatial layouts for understanding information: circle maps, double bubble maps, force field diagrams, squid diagrams, concept maps.

Visual Frameworks in Practice: The Circle Map + Double Bubble Map

Viczian demonstrates with the dilemma: “Should I quit my day job to start my own business?”

Circle Map (defining context from different frames of reference)

┌─ Frame: ME (the entrepreneur) ──────────────────────┐
│  ┌─────────────────────────────────────┐             │
│  │  Following my passion               │             │
│  │  Creative control & freedom          │             │
│  │  Being my own boss      ┌────────┐  │             │
│  │  Working long hours     │ Quit   │  │             │
│  │  Risky but exciting     │ my job │  │             │
│  │                         └────────┘  │             │
│  └─────────────────────────────────────┘             │
└──────────────────────────────────────────────────────┘

┌─ Frame: MY WIFE (sharing the risk) ─────────────────┐
│  ┌─────────────────────────────────────┐             │
│  │  Proud but terrified                │             │
│  │  Losing weekends & holidays         │             │
│  │  Higher stress at home  ┌────────┐  │             │
│  │  May need to work more  │ Quit   │  │             │
│  │  Unpredictable schedule │ my job │  │             │
│  │                         └────────┘  │             │
│  └─────────────────────────────────────┘             │
└──────────────────────────────────────────────────────┘

Double Bubble Map (comparing perspectives)

        MY VIEW                SHARED              WIFE'S VIEW
    ┌──────────┐          ┌──────────┐          ┌──────────┐
    │ Creative │          │Hard work │          │ Forced   │
    │ freedom  │──────────│  Change  │──────────│lifestyle │
    └──────────┘          │High stress│         │ changes  │
    ┌──────────┐          │Fulfilled │          ┌──────────┐
    │  My own  │──────────│   life   │──────────│ Losing   │
    │   boss   │          └──────────┘          │ freedom  │
    └──────────┘                                └──────────┘

The power: by mapping both frames of reference visually, you empathize with the other party and see where differences actually lie — something linear text makes much harder.

The 4 Dimensions of Visual PKM

Viczian’s evolution model of knowledge management:

In 4D Visual PKM using Obsidian Excalidraw:

• Each note is a visual canvas first (front of the postcard)
• Text is on the back (flip the page for metadata and writing)
• Every visual element can be a link to another note
• You navigate your entire knowledge system through visuals

“It’s not that text first and visuals are illustrations — it’s visual first, and you add text to add additional metadata.”

How LearnAI Team Could Use This

• Course design — use circle maps to define learning objectives from student vs. instructor frames; double bubble maps to compare approaches
• Research synthesis — lay out paper contributions visually instead of linear literature review notes; the Game of 15 lesson applies directly to research: the right representation reveals solutions that text hides
• Student assignments — have students create visual summaries instead of essays; tests comprehension more deeply than text reproduction
• AI literacy — teach students to recognize when AI is “completing their sentences” vs. when they’re doing original thinking
• Meeting notes — replace linear meeting minutes with spatial layouts that show relationships between decisions

Real-World Use Cases

• Decision making — circle map + double bubble map for any multi-stakeholder decision (career changes, project priorities, team conflicts)
• Book notes — Viczian’s “Book on a Page” method: visual summary of an entire book on one canvas, with linked detailed notes
• Presentation design — this entire talk was built in Obsidian Excalidraw with every element linked to source notes
• Teaching complex topics — the Game of 15 → Tic-Tac-Toe transformation is a masterclass in how representation changes understanding
• Personal reflection — WYSIATI awareness: before making decisions, ask “what am I NOT seeing?” and use visual exploration to find it

Tools & Resources