The irreducible epistemic atoms underlying the curriculum. 4,828 atoms across 8 types and 2 molecules
Declare your specific intention before encountering competing demands to pre-load attentional filtering without ongoing willpower expenditure.
Conduct periodic authority audits by listing every source shaping beliefs in a domain and evaluating each against basis of trust, scope of expertise, and verification recency, because authority delegations accumulate unconsciously and drift without detection.
When large language models express 90%+ linguistic confidence, independently verify claims in consequential domains, because the gap between expressed confidence and actual accuracy represents systematic miscalibration that human trust heuristics cannot detect.
Form your own judgment before consulting AI systems, then compare rather than defer, because maintaining independent assessment as a non-negotiable baseline prevents automation complacency and preserves epistemic authority.
Calibrate your internal authority voice by recording past predictions with confidence levels, checking outcomes, and adjusting future confidence based on your actual track record rather than emotional intensity or social pressure.
When AI outputs contradict your examined analysis, evaluate them by asking whether they present unconsidered evidence or identify verifiable reasoning errors, treating model fluency as orthogonal to epistemic authority.
Practice daily sovereignty audits by identifying one belief you acted on, tracing its origin, and evaluating whether you would choose it independently, converting self-authority from occasional act to dispositional tendency.
Detect self-authority decay by monitoring for stopped disagreement, adopted positions without reasoning trails, automatic deference, and boring sovereignty journal entries, treating absence of epistemic friction as warning signal not achievement.
Examine past peak experiences, emotional responses, and imagined futures (counterfactual scenarios) as three independent evidence streams that triangulate toward your actual values with greater reliability than introspection alone.
Decompose each peak experience into specific conditions (role, challenge level, autonomy, social context, stakes) rather than accepting surface-level activity labels, because values reside in the conditions of engagement rather than the activity itself.
When resentment persists beyond the immediate moment, extract the violated value through a four-step protocol (notice, identify trigger, name violated standard, record) before either suppressing or indulging the emotion.
Pre-commit to decisions during moments of clarity to eliminate choice points during moments of vulnerability when hyperbolic discounting predictably reverses preferences.
When stated values diverge from revealed values, treat the gap as a design problem requiring infrastructure changes rather than a character problem requiring willpower, since willpower without redesign merely patches symptoms.
Distinguish peak experiences (moments of deep engagement regardless of external validation) from peak achievements (impressive outcomes) when analyzing values, since achievements reveal what your culture rewards while experiences reveal what you intrinsically value.
Use the 'five whys' technique on any significant energy investment to trace instrumental values back to terminal values, stopping only when the answer is 'that's what a good life means to me' rather than pointing to something beyond itself.
Test whether a value is truly terminal by imagining its primary instrumental path disappeared—if you would find alternative paths to the same end, the value you named is instrumental; if the loss feels like loss of meaning itself, you've confused the instrument with the destination.
Run continuous alignment checks asking: (1) Can I trace this activity to a core value in 2-3 steps? (2) Has this instrumental value become my identity? (3) Would I choose differently if this instrument disappeared?
Accept that honoring one value often requires sacrificing another, and interpret the emotional discomfort of that sacrifice as accurate information about real cost rather than evidence you chose wrong.
Use pairwise comparison between values rather than top-down ranking to reveal your authentic hierarchy, because comparing two at a time bypasses the desire to rank values in socially desirable order.
When articulating values, include a disqualification test—a specific condition that would prove the value is aspirational rather than operative—to distinguish values you hold from values you wish you held.
Test stated values through escalating hypothetical trade-offs that increase the cost until the answer becomes genuinely uncertain, because only trade-offs that produce discomfort reveal operative hierarchy.
Use values to eliminate unviable options rather than to identify a single correct choice — a functional values system should narrow your decision space without collapsing it to one option.
Set a single prioritized intention the evening before work begins to eliminate cognitive startup cost and decision overhead when attentional competition is highest.
Treat moral injury—sustained violation of deeply held values—as a distinct diagnostic category requiring values realignment rather than stress management or increased willpower.