Sources: 2026-08 / week-4

Claim ledger for "Why Triangulating Your Assessments Never Closed the Performance Gap."

Claim (as written)	Source	Link	Status
Typical performance (ordinary, sustained, unobserved conditions) and maximum performance (high-stakes, full-effort conditions) are distinct and "correlate only weakly."	Sackett, Zedeck & Fogli, "Relations Between Measures of Typical and Maximum Job Performance," Journal of Applied Psychology 73(3), 1988, 482-486	https://experts.umn.edu/en/publications/relations-between-measures-of-typical-and-maximum-job-performance	`verified`
Ordinary conditions run largely on the brain's memory systems (stored, practised responses); high-demand, novel conditions hand control to other brain systems. Assessment of typical behaviour is silent about the systems that decide the hard case.	Vanaya NeuroTech, Module 01, Neurometric Essentials v1.0, sections 2.1-2.2 (memory system vs the parallel architecture of executive, emotional-regulation, attention, social and motivational systems)	Internal document	`internal`
Triangulation reduces error only across independent sources; agreement among instruments sharing a layer is not independent confirmation. Existing assessment instruments read the observable midstream; capacity sits upstream of all of them.	Vanaya NeuroTech, Module 01, sections 1.2, 2.2, 3.1 (upstream/midstream model; instruments read behaviour, preference, perception)	Internal document	`internal`

Handling notes

Direction chosen: the layer argument, not the method-independence argument. An earlier outline leaned on common-method variance (the instruments are not statistically independent). Twin redirected to the on-brand version: the instruments may be perfectly valid and genuinely different in method, yet all sample the midstream, so triangulating them cannot reach the upstream. This is more consistent with the month and carries less villain risk, since it does not impugn the tools' validity. The independence point survives only as the plain observation that shared-layer agreement is not independent confirmation, which the article makes without a statistical claim.
Sackett et al. sampled supermarket cashiers. The paper establishes the typical-vs-maximum distinction and its weak correlation in that setting. The article cites it for the distinction itself, not for an executive-specific magnitude. The neural mapping (memory system for typical, other systems for maximum) is the module's framing and carries the leadership relevance; logged internal.
Villain control. The "boundary of the instrument, not a failure of it" section is load-bearing and deliberately generous: each instrument is named as valid for its purpose. No named firms, no named instruments, comparisons rendered as categories only. The module's internal comparator tables were not reproduced.

Week 1 owns "you fund contents"; week 4 owns "you measure contents." Week 4 does not reuse week 1's experience-does-not-predict finding (Van Iddekinge). Its fresh anchor is the typical-vs-maximum distinction, which is distinct from every prior week's citation (cognitive load, fluid vs crystallized, stress-to-habit).

Common-method-variance literature (Podsakoff et al. 2003) was considered for the independence argument and dropped with that direction. Available if a future, methodology-focused piece ever wants it.