EXP-DT-BOOTSTRAP

… pending

Developmental Traversal Bootstrap — calibration before full run

2026-04-16 L0-L6 level paper3 10 runs $0.06

new

premortem

mock

real

metrics

analyze

review

krit

done

In plain language

10 short replicates to estimate variance before committing $5 to the full experiment.

▶ Technical details

Research hypothesis

Bootstrap calibration: 10 × 100-round mini-replicates with minimal agents under scarcity. Measure level-detection variance across replicates. Determine required N for statistical ordering test (Lenski: need enough replicates to detect ordering consistency). If variance low → N=3 full replicates sufficient. If high → need N=10+.

Experimental setup

Type: multi_condition

Condition	Parameters
DEV	label: Minimal agents, T=0.7, budget=300 (100 rounds), interaction: LIVE, persona: IDENTICAL, n_runs: 10

Parameters

n_agents: 5
n_rounds: 100
model: google/gemini-2.5-flash-lite
temperature: 0.7

developmental-traversal bootstrap calibration phase-2