Initial Commit

experiments/002_conversion_breach/hypothesis.md

@@ -0,0 +1,34 @@
+# Experiment 002 Hypothesis
+
+## Primary hypothesis
+
+Direct control plus practical stakes will turn systems knowledge from implementation work into personal agency. The player will care about understanding because it produces survival, power, improvisation, or recoverable chaos.
+
+## Controlled comparison
+
+- **Calm:** constructs have the same properties and physics but do not pursue or damage the player.
+- **Breach:** constructs pursue, collide with, and damage the player; waves continue.
+
+## Competing explanations
+
+1. Direct systemic manipulation is interesting in both modes; pressure changes pacing but is not necessary.
+2. The system becomes meaningful only under pressure.
+3. Pressure interrupts experimentation; calm is more promising.
+4. Action feedback is momentarily stimulating but the laws remain shallow.
+5. The mechanics feel like ordinary combat abilities rather than magical engineering.
+
+## Strong evidence
+
+- A discovery produces a second voluntary test.
+- The player deliberately engineers a state before acting: brittle, unstable, massive, aligned, or grouped.
+- A chosen calibration supports a self-selected tactic.
+- The player recovers from a bad state using the system rather than merely kiting.
+- The calm/breach preference has a clear reason tied to reasoning or emotional stakes.
+
+## Weak or negative evidence
+
+- The player spams one verb without tracking properties.
+- Conventional movement and aiming do all the enjoyable work.
+- Temperature states are merely status effects with obvious prescribed combos.
+- Calibration is generic numerical progression without changing decisions.
+