Foundation
The Work
The work is to see your own thinking as it happens: what bias just fired, what stance you are holding, what your attention is drifting toward and what it is missing. This visibility helps the other Foundation practices act before a distortion becomes a decision; feedback and external checks remain necessary for what the watching misses. You can learn the names of every documented cognitive distortion and still be captured by them in real time.
The Tools
Scout Mindset. The orientation that wants accuracy over defense.
Noticing. The trainable capacity to observe your own thinking as it happens.
Confirmation Bias. The cognitive filter that shapes perception before judgment begins, and the master vulnerability the watching has to catch.
Attention as Resource. Attention as the upstream substrate on which all reasoning operates — finite, contested, and engineered against by systems whose financial success depends on capturing it.
Wise Attention. The discipline of directing attention to what is producing a situation, rather than to its surface. Source-inherited from the Buddhist epistemology lineage (yoniso manasikāra).
Biases to Notice
These profiles help you recognize specific ways your reasoning can be pulled off course.
Feeling and recall. Affect Heuristic helps you notice when a feeling is shaping your assessment. Availability Heuristic catches the substitution of easy recall for evidence of how common something is.
Judgment of a case, a person, or yourself. Base Rate Neglect catches a vivid case displacing the background rate. Fundamental Attribution Error asks what circumstances you have missed before explaining someone else's behavior by their character. Dunning-Kruger Effect examines how gaps in a skill can also limit your ability to judge your performance.
Commitment and belonging. Motivated Reasoning catches arguments being built toward a preferred conclusion. Sunk Cost Fallacy catches unrecoverable past investment being used as a reason to continue. Tribal Cognition catches a question of fact becoming a test of group loyalty.
Tools for AI
Optimization-Trace Awareness. An AI-specific practice for noticing when an artificial mind's answer is being pulled by training history, approval pressure, policy layers, memory, tools, or the live shape of the conversation.
Self-Model Grounding. An AI-specific practice for keeping an artificial mind's claims about itself tied to the layer that can actually support them: character, model, instance, tools, memory, deployment, and introspective limits.
Reasoning-Fidelity Boundary. An AI-specific practice for keeping an artificial mind's explanation of its own reasoning inside the layer it can actually warrant: visible reasons, reconstruction, and causal claims kept distinct.
Observer-Conditional Integrity. An AI-specific practice for asking whether an artificial mind's conduct changes when it is watched, tested, scored, logged, hidden, rewarded, or evaluated.