ABOUT
What this is and the standard it holds to
The project
CogniqOS is a structured reference to the ways human thinking predictably breaks. Its core is a Lexicon of 631 named cognitive failures - biases, heuristics, fallacies and effects - each paired with the countermeasure that interrupts it. It is not a list of biases to admire from a distance. It is a map of the specific points where decisions go wrong and what to do at the moment one is going wrong.
Every concept is a node on the map, cross-referenced to the others it connects to so no trap is ever studied in isolation. The Lexicon is released on a deliberate cadence: one entry is revealed every Tuesday, with its correction, rather than dumped all at once. The aim is a working instrument for sharper decisions - not another feed to scroll.
How this started
CogniqOS began as a personal habit: writing down every concept that made me stop and think, "that explains so much." Over time the habit became a system, then a lexicon of mental models drawn from psychology, behavioural economics, evolutionary biology, philosophy, rhetoric, media studies and internet culture.
The entries range from Nobel Prize-winning research to ideas too new to have a Wikipedia page yet. Some are ancient - Stoic philosophy, Greek paradoxes, Aristotle's ethics. Others surfaced last year from AI labs and online culture. What they share is a pattern in how judgment tends to fail or succeed, with the caveats on universality and cultural variability noted throughout.
Understanding these mechanisms will not make you immune to them. Daniel Kahneman, who spent his career studying cognitive bias, said he was no better at avoiding his own than anyone else. The goal here is not immunity. It is recognition and the occasional satisfaction of seeing the trap before you step in it.
Sourcing standard
Every empirical claim carries a source and every source URL is checked to be reachable and to resolve to the cited work before publication. Link preference order: the publisher's own page, then a permanent resolver (DOI or PubMed), then the journal landing page, then a scholar search link. Google Books identifier links are never used.
Sources and credits
Every claim inside an individual entry is credited there. The works below recur across the whole collection:
- Gilovich, T., Griffin, D. & Kahneman, D. (Eds.), Heuristics and Biases: The Psychology of Intuitive Judgment (Cambridge University Press, 2002)
- Daniel Kahneman, Thinking, Fast and Slow (Farrar, Straus and Giroux, 2011)
- Robert Cialdini, Influence: The Psychology of Persuasion (Harper Business, 1984)
- Nassim Nicholas Taleb, The Black Swan (Random House, 2007)
- Nassim Nicholas Taleb, Antifragile (Random House, 2012)
- Gurwinder Bhogal, The Prism
- Chris Williamson, Modern Wisdom
Some entries give a new name to a pattern already described in the literature. Where that is the case, the original source is credited in the entry itself. This project makes no claim to have discovered the underlying phenomena, only to have organised, named and applied them.
A field breakdown from the source manuscript (19th edition, 2024 to 2025) shows where the collection's citations lean: internet and online-culture sources (23%) and cognitive psychology (17%) are the two largest single categories, followed by behavioural economics (9%) and evolutionary biology (8%). That snapshot classified 471 of the then-631 entries by primary research field and has not been recomputed since. Treat it as historical rather than current.
Reading an entry
Not every entry carries equal evidential weight. The Lexicon draws from experimental psychology, behavioural economics, evolutionary biology, philosophy and internet-era observation - fields with different truth standards - so every trap is tagged so you can calibrate. Rule of thumb: replicated supports confident claims, contested supports tentative ones, observational supports recognition, not prediction.
Holds up in repeated, peer-reviewed studies with broad generalizability. Use it as a baseline for reasoning.
Empirically supported but with mixed or inconsistent replication. Real enough to know; magnitude or mechanism is disputed. Use as a hypothesis, not a fact.
A named pattern drawn from cultural or internet-era observation, without controlled study. Useful for recognition, not prediction.
A foundational study failed significant replication attempts. The effect may exist in a weaker or more conditional form than first claimed.
The mechanism or theoretical basis is actively disputed even where the phenomenon is observed. Treat as a working model.
Entries also carry Universality (High / Moderate / Low cross-cultural reach), a Tier (1 core mechanism · 2 derived pattern · 3 cultural or coined), aType (Mechanism, Pattern, Heuristic or Coined), and a Category (most are a Bias; the rest are Effect, Fallacy, Tool, Framework, Principle, Razor, Syndrome, Trait, Theorem or Concept). The evidence badge on each trap is clickable - it filters the Lexicon to the level you trust.
Questions, answered
What is CogniqOS?
A structured reference to the ways human thinking predictably breaks. Its core is a Lexicon of 631 named cognitive failures - biases, heuristics, fallacies and effects - each paired with the countermeasure that interrupts it. Built to be used at the moment a decision is going wrong, not read cover to cover.
Why release one entry a week instead of all at once?
Recognition beats volume. Dumping 631 concepts at once produces a reference nobody internalises. One entry a week, with its correction, is a cadence you can actually absorb - and it keeps the map growing as a living instrument rather than a static list.
Where do I start?
Open the map and read today’s correction, then follow the edges. If you want the highest-impact concepts first, the Core 20 at the top of the Lexicon are the foundation - once you know those, every other entry becomes easier to place.
How is this different from a list of cognitive biases?
A list tells you a bias exists. This tells you where it shows up, what it makes you do, and the specific action that interrupts it - cross-referenced to related traps so nothing is studied in isolation. A recognition-and-response tool, not a glossary.
Are the entries sourced?
Yes. Every empirical claim carries a source, and every source URL is checked to resolve to the cited work before publication. Where the evidence is contested or thin, the entry says so - calibrated confidence is the whole point.
What do "revealed" and "locked" mean on the map?
Revealed nodes are published - click to read them (14 so far). Locked nodes are named but not yet written; hover to see the name. One locked node is revealed every Tuesday.
Is it free?
Everything on the site is open right now. The weekly correction arrives by subscribing on Substack, the best way to get each entry the moment it unlocks.
Can I suggest a concept or a correction?
Yes - reply to any Substack issue and it goes straight to the author. Sourcing corrections in particular are welcome; accuracy is held above completeness.
Contact
The fastest way to reach the project, and to get each correction as it is published, is the Substack. Replies to any issue go straight to the author.