Co-Cognition Map

🚧 Coming Soon

Human cognition and artificial intelligence do not exist in isolation. Across every domain of thought — from creative writing to mathematical proof, from ethical reasoning to pattern recognition — human minds and LLMs engage in an intricate, ever-shifting dance. The Co-Cognition Map is Co-Cognition Lab's answer to a deceptively simple question: What does the full landscape of human-AI cognitive collaboration actually look like?

What Is the Co-Cognition Map?

The Co-Cognition Map (Project 2) is a cross-disciplinary domain taxonomy designed to systematically chart the terrain where human cognition and LLM capabilities converge, diverge, and complement one another. Building on the theoretical foundations laid by Project 1 — our work on LLM Intuition and the Complementarity Map — this project transforms abstract insight into a practical, navigable framework.

Unlike existing taxonomies that treat AI capabilities or cognitive tasks in isolation, the Co-Cognition Map is built from the ground up to capture interaction dynamics: how human intuition and LLM processing trade off, reinforce, and sometimes interfere with each other across different types of problems.

The Taxonomy Framework

At the heart of the Co-Cognition Map lies a multi-dimensional classification system. Each cognitive domain is evaluated along carefully designed scoring dimensions that capture:

• Complementarity potential — Where do human and AI strengths combine to exceed either alone?
• Autonomy risk — In which domains does heavy LLM use risk constitutive degradation of human capability?
• Collaboration readiness — How mature are the interfaces for genuine co-cognition in this space?
• Intuitive load — How heavily does the domain rely on the kinds of fast, non-verbal cognition that our LLM Intuition research has mapped?

This scoring system enables a nuanced view: not simply "AI can or cannot do this," but rather "what kind of partnership is possible here, and what are the conditions for making it fruitful?"

Planned Content Modules

Taxonomy: The Classification System

The cross-disciplinary classification itself — a hierarchical, multi-dimensional taxonomy covering the full breadth of cognitive domains where human intuition and LLM capabilities intersect. Each domain is tagged with its scoring profile, creating a data-rich map of the co-cognitive landscape.

Domain Annotations

Detailed, research-grounded analysis of individual cognitive domains. Each annotation examines the specific character of human-AI interaction within that domain: where human judgment remains indispensable, where LLMs offer genuine augmentation, and where the boundaries between the two are most productively porous.

Trading Zone Analysis

A systematic exploration of trading zones — the boundary regions where human cognition and AI capabilities meet, negotiate, and co-produce outcomes. Drawing on Peter Galison's concept of trading zones in science, we map how these cognitive borderlands function: what gets traded, what gets lost in translation, and how fluent co-cognition can be cultivated across the boundary.

Exploration Roadmap

A guided, systematic approach to navigating the full cognitive landscape. The roadmap offers prioritized pathways through the taxonomy — entry points for researchers, practitioners, and teams seeking to understand where and how to deploy human-AI collaboration most effectively.

Development Status

Status: Taxonomy framework v0.2 complete, external feedback received, domain annotation not yet started.

Expected: June–July 2026 (after framework finalization)