Working Theory v0.21 Preprint PublishedSame Mind Project · Protocol Phase
The Understanding Lab · Independent Research Program · Est. 2026

An independent research program investigating the nature, measurement, and development of understanding across humans and intelligent systems.

Scroll to explore
The Question
What does it mean to truly understand something?

Conventional assessment can tell us whether an answer is correct. But correctness alone may not reveal the structure, stability, transferability, or limits of the knowledge that produced it.

A person can produce a correct answer through memorization, guessing, imitation, or deep conceptual comprehension. These are fundamentally different cognitive processes — yet they frequently produce the same observable output.

The core problem
Memorization≠Understanding
Pattern matching≠Understanding
Guessing≠Understanding
Imitation≠Understanding
...but all can produce the same observable output: a correct answer.
Current Theory
Working Definition · v0.2 · September 2026

"Understanding is an organized, relational capacity through which an agent can make meaningful use of what it knows, including by explaining, reasoning, predicting, transferring, adapting, or revising its knowledge when relevant conditions change."

This is a working definition, not a final scientific claim. Revised after a formal adversarial stress test in September 2026.
A crucial three-level distinction:
WHAT IT IS
Understanding
↓
HOW IT MAY MANIFEST
Explanation · Reasoning · Prediction · Transfer · Adaptation · Revision
↓
HOW WE OBSERVE IT
Tasks · Responses · Behavior · Confidence · Consistency
These levels must not be collapsed. A measurement framework is strongest when it makes their relationships explicit.
Research Architecture

How the Lab investigates understanding

The Lab treats theories as things to test, not beliefs to protect. Every project connects to the same cycle — from question to revision.

HCMS is one instrument within this architecture. The Same Mind Project is the first empirical test. Neither is the destination.

revision loopQUESTIONThe founding questionTHEORYv0.1 frozen · v0.2 workingTESTAdversarial + empiricalEVIDENCESame Mind Project →MEASUREMENTHCMS instrumentREVISIONTheory evolves with evidence↺ theory revises when evidence demands it
Research Programs

Active and planned investigations

EXP-001
Human Cognition Measurement System
Preprint

The first experimental instrument. Explores whether multiple observable signals can provide a more informative picture than correctness alone. DOI: 10.5281/zenodo.18269740

EXP-002
Same Mind Project
Protocol Phase

Can people who answer the same question correctly demonstrate different structures of understanding? The first empirical study.

EXP-003
AI Audit #1
Future Research

Can the Same Mind framework apply to AI systems? Do AI systems exhibit the same observable patterns as human understanders?

The question remains open.

The Lab does not claim to have solved the problem of understanding. It exists to make the problem more precise, test possible answers, discover where they fail, and gradually build better ones.