How the lab finds and frames a problem.
The lab's method merges three established practices to surface research directions that are at once genuinely novel and reasonable to pursue, then adds a requirement of its own.
Literature-based discovery (after Swanson) finds connections between fields — or between specialties within one field — that do not cite each other. This is where a pairing nobody has made comes from.
Morphological analysis (after Zwicky and Ritchey) maps which combinations of known results have never been tried together, turning a loose connection into a structured, checkable candidate.
Strong inference (after Platt) admits only questions a decisive test can settle, filtering out the merely clever.
Novelty comes from the first, a tractable test from the third, and the second connects them.
To these the lab adds a requirement of its own: no candidate is pursued until it is anchored to a fundamental limit — a hard floor traced to a theorem, conservation law, or measurement bound. The three practices are established and credited as such; the assembly and that floor are the lab's own.
Work is run to a few standards regardless of subject: pre-registered where possible, with confirm, refute, and inconclusive rules fixed before data is examined; artifacts hashed so results can be checked against what was actually run; and precision and noise limits stated rather than omitted. A negative result is treated as a result.