On the nature of
systematic error.

We are interested in the gap between what minds are capable of and what they actually do — and in the structural conditions that make that gap predictable.

Cognition and its limits

Human cognition is not broken. It is optimised — for environments that no longer fully exist. The heuristics that allow fast, low-cost inference under uncertainty also produce systematic, reproducible error when conditions change. This is not a flaw to be corrected; it is a feature of any system that must act faster than it can deliberate.

What interests us is the architecture of that error: why particular biases cluster, why they interact, and why the same distortions appear independently across cultures, disciplines, and historical periods. The recurrence is the signal.

Metacognition as instrument

Knowing that a bias exists does not reliably reduce its influence. This is itself a documented phenomenon. The gap between conceptual knowledge and behavioural immunity is one of the more uncomfortable findings in the literature — and one of the most useful to sit with.

We are less interested in checklists than in the structural conditions that make reflective thought possible or impossible. Context, incentive, time pressure, social environment: these determine far more than individual intention does.

Scope

Our interest spans cognitive psychology, behavioural economics, philosophy of science, statistical epistemology, and organisational theory — not as separate disciplines but as different angles on the same underlying question: how do reasoning systems, human or institutional, fail in structured ways?

The Atlas is one attempt to map that territory. It will not be the last.

On method

The tools we build are deterministic and citable. We are skeptical of approaches that treat ambiguity as a feature. The goal is not neutrality — neutrality is its own epistemic position — but precision about what is known, what is contested, and what is conjecture.

This site is designed to survive trends.

Support

Incognati is free and will stay free. There are no ads, no paywalls, and no tracking beyond anonymous visit counts.

If you find it useful and want to help keep it running, a small contribution is appreciated — but entirely optional.