The irreducible epistemic atoms underlying the curriculum. 4,828 atoms across 8 types and 2 molecules
Evaluate source credibility before consuming content, not after, because evaluating post-consumption cannot recover the cognitive resources already spent processing noise.
Follow curators with demonstrated track records rather than consuming primary sources at scale, because effective curators compress signal-to-noise ratios across entire domains through pre-filtering.
Limit information sources using one-in-one-out rules, because unbounded source accumulation guarantees eventual overload regardless of source quality.
Capture and progressively summarize consumed information to transform it from passive consumption into retrievable thought-objects.
Set fixed time budgets for information consumption independent of available material, because unbounded consumption time allows noise to expand indefinitely regardless of signal availability.
When the volume of available information exceeds cognitive processing capacity, prioritize stopping acquisition over increasing processing — saying no to information sources produces better outcomes than attempting to read faster or more efficiently.
Use satisficing decision rules (define 'good enough' thresholds and stop searching when met) rather than maximizing strategies when information is abundant and time-constrained, because maximizing produces worse subjective outcomes despite objectively equivalent or better choices.
Build deep expertise in 2-3 signal-critical domains before expanding breadth, because domain-specific perceptual structures created through deep engagement enable signal detection that surface-level familiarity across many domains cannot produce.
Measure both process metrics (daily actions) and outcome metrics (end results) in paired systems, using process metrics for real-time adjustment and outcome metrics for validating that process metrics remain predictive.
When a measure becomes a performance target, immediately establish a paired outcome metric that the leading indicator was meant to predict, because optimization pressure severs the causal link between indicator and outcome.
Implement a minimum one-hour delay between encountering emotionally provocative information and evaluating or acting on it, because hot-state judgments systematically differ from cold-state judgments in ways that degrade decision quality.
Label emotional reactions verbally and specifically before evaluating the information that triggered them, because affect labeling activates prefrontal regulatory circuits and reduces amygdala reactivity, converting reaction into observation.
Use cognitive reappraisal (reinterpreting the meaning of an emotion-triggering event) rather than expressive suppression (hiding emotional reactions) when managing emotional responses to information, because reappraisal reduces internal emotional experience while suppression increases physiological stress without reducing the emotion.
Combine multiple imperfect upstream metrics into composite leading indicators rather than relying on single predictors, because composite signals smooth individual noise while preserving shared directional information.
Break information transmission chains by moving closer to primary sources when decision stakes justify the cost.
Design external capture systems to preserve contextual detail that verbal summaries eliminate.
Implement retrieval practice rather than repeated study to distinguish actual understanding from recognition fluency.
Schedule incubation periods with low cognitive load to enable default mode network consolidation after intensive learning or problem-solving.
Use information removal (fasting) as a diagnostic tool to reveal genuine needs versus habitual cravings before adding new inputs.
Tag captured information with estimated decay rates and implement scheduled depreciation reviews to prevent outdated knowledge from corrupting decision-making.
Implement a connection-rate audit where every consumed piece of information must explicitly link to existing knowledge or be classified as non-learning consumption.
Set detection thresholds based on the relative costs of false positives versus false negatives rather than maximizing either sensitivity or specificity alone.
Train perceptual systems through structured pattern exposure rather than through explicit rule-following or information accumulation.
When evidence for a decision is ambiguous and new information is arriving, preserve optionality by delaying commitment rather than forcing resolution.