EXP-DOSE40-SINK

informative

Integrative Dose-Response — 40% Pure SINK (2/5 Synthesizers)

2026-03-09 L2 level paper3 1 runs $0.01

new

premortem

mock

real

metrics

analyze

review

krit

done

In plain language

In the previous experiment, two different "integrative" agents did not suppress hierarchy — one absorbed vocabulary (synthesizer), the other amplified flow (curator). Here we deploy TWO synthesizers to check: is it the type of integration that matters, not the number of integrators?

What we found

No predictions scored

Predictions we made before running

0/4 confirmed

▶ Technical details

Research hypothesis

SINK fraction, not generic 'integrative fraction', is the operative variable for hierarchy suppression. Two pure SINK-type agents (synthesizers) at 40% will suppress hierarchy MORE than the mixed SINK+THROUGHPUT condition (EXP-DOSE40 TR=0.437), approaching EXP-ENGINEER levels (TR=0.240) or below.

Experimental setup

Type: single_condition

Condition	Parameters
PERSONA_LIVE	persona: PERSONA, interaction: LIVE

Factors: persona (PERSONA) × interaction (LIVE)

Parameters

n_agents: 5
n_rounds: 50
n_runs_per_condition: 1
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).