March 4, 2026Pillar Report
Human Performance
Medium Confidence4 signals / 15 sources
Human performance optimization is progressively colonizing and eroding the cognitive substrates it depends on.
Signals
A
Fowler/Morris: humans own loop structure, not outputs.
New cognitive demand class most workers not trained for.
A
LOMS announced. No methodology or peer review indicated.
Provider-authored assessment inverts scientific epistemology.
B
Nguyen argues quantified scoring replaces intrinsic value with metric pursuit.
Gamification as cognitive colonization. Telos erosion.
Control Surfaces
| Lever | Status | Change | Evidence |
|---|---|---|---|
| Human judgment in AI loops | Contested | Delegation pressure increasing | Fowler/Morris |
Watchlist
- ConfirmationIndependent AI tutoring longitudinal studies
- InvalidationAI learning improves deep reasoning over 24+ months
Falsifiers
- AI learning improves reasoning transfer over 3+ years
- Loop architect cognition widely teachable in 12-18 months
Key Unknowns
- LOMS peer review status
- Effect sizes of AI on developing cognition
Noise Filter
- Methamphetamine as nootropic— Zero analytical value
Get The Signal daily
Cross-domain structural analysis, delivered every morning.