EXP-RNSH-G3

informative

Ratchet Null (shuffled-input null model) — G3

2026-03-27 L6 level paper3 1 runs $0.00

new

premortem

mock

real

metrics

analyze

review

krit

done

In plain language

This experiment tested a simplified version of the system by shuffling the input data to see if the original system's performance was due to chance. However, the experiment was not able to produce meaningful results because it was run with too few agents. Therefore, no conclusions could be drawn about whether the system's performance was due to chance or genuine learning.

What we found

UNDERPOWERED (N=1): No predictions scored

Predictions we made before running

0/1 confirmed

▶ Technical details

Research hypothesis

G3 of shuffled-input null model. 3 agent(s) × 10 rounds.

Experimental setup

Type: single_condition

Condition	Parameters
G3_NULL	persona: PERSONA, interaction: LIVE

Factors: persona (PERSONA) × interaction (LIVE)

Parameters

n_agents: 3
n_rounds: 10
model: gemini-2.5-flash-lite
temperature: 0.9
scheduler: round_robin

Trophic Ratios by Condition

Mean trophic ratio per agent across runs. Error bars = ±1 std dev. Higher TR = more upstream (exporter).