Magnetic Regression Index

VARIATION

abstraction

Path A — Regression

ONE ANSWER BECOMES THE RULE

Path B — Discovery

changed conditions

A STRUCTURE THE ORIGINAL MEASUREMENT COULD NOT REVEAL

A conceptual diagram, not an experimental result. Abstraction can compress variation into a single answer that becomes a rule. MRI asks whether that answer still holds when the conditions that produced it change. A different result does not by itself prove Magnetic Regression.

Every measurement records an outcome under particular conditions. What happens when the conditions change?

MRI compares defensible views, changes conditions where possible, and tests what remains. The question is not whether the past was wrong. It is whether an earlier description remains decision-relevant when the world that produced it changes.

Explore magnetic regression

The investigations

MRI-001 47 to 1 A different definition of work changes the model ranking. Not a different answer to the same job, but a different job. MRI-002 Category Capture How competing meanings of a category relate to recognition in New York Times discussions. MRI-003 · Local prototype The Best Specimens Inspect the propositions in the 1904 discussion and the construction of Galton’s numerical evidence, with sources and limits preserved.

Further investigations will appear here.