The irreducible epistemic atoms underlying the curriculum. 2,888 atoms across 3 types and 2 molecules
Separate thoughts from identity by using language that marks them as observable objects rather than self-definitions.
Build composable idea structures by externalizing thoughts as atomic units that can be combined to generate insights no single thought contains.
Optimize for capture latency over capture quality—reduce the time between thought occurrence and externalization to under 5 seconds regardless of tool fidelity.
Externalize all open commitments into a trusted system to close Zeigarnik loops, as formulating specific plans (not completing tasks) resolves cognitive tension and frees working memory.
Use externalization as a diagnostic tool for understanding gaps—write to discover what you don't know rather than to record what you do.
Structure knowledge artifacts with progressive layers of refinement where each layer demands increased cognitive engagement and produces sharper understanding.
Cultivate the observer stance by explicitly naming your current cognitive state during decision-making to enable real-time metacognitive regulation.
Before committing to any plan, assume it has already failed and generate specific causal mechanisms of failure to surface risks that optimism bias suppresses from standard risk assessment.
Build external inventory systems to compensate for context-dependent retrieval rather than trusting mental completeness.
Use base-rate forecasting from similar past cases rather than bottom-up task simulation to counteract systematic underestimation of duration and complexity.
Implement complete externalizations (mind sweeps) periodically to surface commitments and concerns that current context has suppressed from active recall.
Structure externalized knowledge with sufficient metadata and linkage that AI can traverse your full corpus to surface connections you cannot retrieve in working memory.
Train metacognitive monitoring and control as separate, deliberate skills rather than waiting for spontaneous self-awareness to emerge.
Treat beliefs about skill malleability as directly trainable rather than fixed personality traits, because those beliefs determine whether skill-building effort will be invested.
Separate capture from organization into distinct cognitive operations that never overlap, because working memory cannot handle both classification and preservation simultaneously.
Layer organization progressively over time based on actual usage patterns rather than attempting complete up-front taxonomy design.
Deploy AI to handle information logistics and context management while reserving human directed attention for evaluation, synthesis, and judgment to prevent attentional infrastructure atrophy.
Externalize cognitive material only to systems you trust and review regularly, because the brain maintains background processes for commitments to untrusted destinations.
Capture context-dependent insights immediately at their point of origin, because environmental and cognitive state changes make reconstruction impossible rather than merely difficult.
Tag captured thoughts as 'signal' or 'narration' during review to train recognition of the distinction and improve real-time filtering over time.
Establish automatic capture behavior through environment-anchored tiny habits before designing organizational systems, because systems cannot function without reliable material flow.
Minimize time-to-capture below three seconds for spontaneous thoughts by using voice-based or always-accessible tools, because friction at capture determines what material enters your cognitive system.
Build the keystone behavior first and allow downstream practices to emerge organically from the existence of captured material rather than designing comprehensive workflows upfront.
Externalize mental models as spatial diagrams rather than prose to force explicit specification of entities, relationships, and structural gaps that language can skip.