Candidate hypotheses · none proven · you decide

Engram Hypotheses

The idea hiding between two sources nobody read together, proposed by a machine, killed-or-kept by a jury, shown with its sources.

Vol. 1 · surfaced by the Engram brain 1 hypothesis · cross-model jury · sourced

Read this first. Every item below is a candidate: an AI proposed it by bridging facts from different sources, and an independent jury of models kept only the grounded, testable ones. They are not discoveries, and we don’t claim to be first to think any of them. Someone may have had the same idea before; what we can show is that this one was reached from grounded facts, not copied from anyone. The value is the question and the receipts: each one links the public sources it was built from, grounded, not invented. You decide what’s worth a lab.

evolutionary biology × economics

Can one information measure clock how fast strategies change in both biology and markets?

evolutionary changeeconomic strategy dynamics · via Fisher information

In evolutionary biology, natural selection has been shown to maximize Fisher information, and the total change selection produces tracks that same information. In economics, strategy frequencies follow the same replicator dynamics, driven by a payoff matrix. The system kept landing on these as one object: in both, the observed acceleration of strategy frequencies encodes an unobserved force, so the Fisher information of a frequency trajectory measures the scale and structure of change the same way in a population and in a market.

✓ jury: kept (Model A) testable · Jun 8, 2026 · read the receipts →