Human Performance
Self-directed optimization is converging on clinical methodology from below, but the felt-sense gap remains the undetected failure mode.
Signals
GRAIL's landmark blood-based MCED trial failed composite primary endpoint. Peter Attia argues secondary data may tell different story.
Technologies that fail broad mandates often succeed as precision tools. Screening paradigm fragmenting from 'test everyone the same' to 'test different people differently.'
Xiong et al. (2025) systematic review + network meta-analysis comparing exercise types and doses for sleep quality improvement.
Shift from 'exercise improves sleep' (directional) to 'which exercise, at what dose, for whom' (prescriptive specificity). Maturation pattern: population truths decomposed into individual protocols.
2026 study examined CoQ10 effects on recovery, stratifying by normal-weight vs. overweight males.
Stratification by metabolic phenotype is the signal. Recovery may need to be prescribed by metabolic state, not just training load.
Discussion examines high-dose stimulant use (60-100mg Adderall), noting arousal past optimal zone degrades complex task performance.
Divergence between felt productivity and measured output is an underappreciated failure mode. Stimulants at high doses push toward simpler work that feels productive.
Community member describes months-long daily logging protocol to assess supplement effects.
Self-trackers converging on single-subject experimental design from below while clinical researchers push n-of-1 trials from above. Gap between biohacker and personalized medicine narrowing at protocol level.
Control Surfaces
| Lever | Status | Change | Evidence |
|---|---|---|---|
| Sleep-exercise dose specificity | Advancing | Network meta-analysis ranks protocols | Xiong et al. 2025 |
| Cancer screening paradigm | Contested | GRAIL trial failure challenges universal MCED | Peter Attia analysis |
Watchlist
- ConfirmationGRAIL releases subgroup data showing cancer-specific detection rates for risk-stratified deployment
- InvalidationFollow-up research shows no subjective-objective divergence at clinical stimulant doses
- ObservableNetwork meta-analyses for exercise proliferate across other outcome variables
Falsifiers
- Studies showing subjective productivity assessments reliably correlate with objective output at high arousal
- Consumer self-tracking produces clinically equivalent outcomes to formal n-of-1 trials
Noise Filter
- NanoJudge LLM ranking tool— AI tooling, not human performance
- GOG (graph-oriented generation)— Software engineering, keyword match on 'memory'
Get The Signal daily
Cross-domain structural analysis, delivered every morning.