The irreducible epistemic atoms underlying the curriculum. 4,828 atoms across 8 types and 2 molecules
Check integrated schemas against their source components to identify what details were dropped during integration, distinguishing between irrelevant details and inconvenient evidence that was excluded to preserve narrative coherence.
Verify that integrated frameworks generate novel predictions or capabilities beyond what component schemas provided independently, as integration that only explains existing knowledge adds narrative rather than understanding.
Delay integration until each component schema can be used independently and correctly, as premature combination of poorly-understood components produces rigid coupling that prevents future schema updates.
Externalize schema comparisons through writing or diagramming rather than attempting integration purely internally, because the number of possible connections grows combinatorially beyond working memory capacity.
Extract the valid insight from each past schema version by understanding what problem it was solving in context, rather than dismissing it as uniformly naive.
Deliberately override your current working self's filtering by actively reaching for memories that supported schemas you no longer hold, because current goals selectively reconstruct which past experiences are accessible.
Prioritize recovering meta-schemas (patterns of thinking, learning strategies, heuristics for uncertainty) from past versions over domain-specific conclusions, because foundational cognitive patterns transfer across contexts while specific conclusions may not.
Design your worldview's protective belt to generate novel predictions rather than only rationalize failures after the fact, using predictive success as a test of whether the worldview is progressing or degenerating.
Revisit and re-engage with foundational knowledge periodically to prevent catastrophic forgetting as new knowledge arrives, using deliberate replay to maintain old understanding while acquiring new.
Conduct integration work as part of daily life rather than as separate intellectual exercise, noticing where understanding fails during action and using lived gaps as integration material.
Recognize that existing habits and automatic reactions are undesigned agents installed by environment and repetition, then audit them before attempting to install new ones.
Offload recurring low-stakes decisions to environmental systems rather than willpower, preserving finite executive function capacity for decisions that require genuine deliberation.
Anchor health behavior triggers to stable external signals rather than subjective internal states, as internal assessments degrade under the conditions when the behavior is most needed.
Track displacement rate rather than perfect execution when replacing default agents, because replacement is a gradual process competing against thousands of prior reinforcements.
Pre-commit recurring decisions to explicit criteria during periods of clear thinking rather than re-deciding each time the situation arises.
Break complex information into chunks of 3-5 items to work within working memory constraints while maintaining functional processing capacity.
Externalize intentions as specific if-then implementation plans to convert goals into automatic context-triggered behavior.
Use specific, measurable goals rather than vague directives ('do your best') — specificity directs attention, calibrates effort, sustains persistence, and activates task-relevant knowledge.
Design behavioral systems to be falsifiable by defining observable success criteria and review cadences to enable improvement through empirical feedback.
Use checklists to delegate memory and sequential verification to external artifacts, freeing working memory for genuine judgment tasks.
Self-monitor behavior through real-time observation and recording to trigger evaluative processes that automatically shift behavior toward personal standards.
Use momentary sampling methods rather than retrospective recall to capture actual behavioral patterns and avoid memory reconstruction biases.
Inventory system components and their interactions before attempting optimization because system behavior emerges from structure not individual component quality.
Prioritize behavioral agents by their firing rate over their sophistication when the two conflict, because consistent execution compounds while sporadic execution decays.