EXP-COMPETENCY-ROTATION-R3

informative

Competency Rotation Replication 3 — rotation=1, seed S4 (direct replication)

2026-04-02 L4 level paper3 1 runs $0.00

new

premortem

mock

real

metrics

analyze

review

krit

done

In plain language

Replication r3 of competency rotation test. Direct replication of original (rotation=1) with different seed. Tests seed sensitivity within same rotation condition.

What we found

UNDERPOWERED (N=1): No predictions scored

Predictions we made before running

0/4 confirmed

▶ Technical details

Research hypothesis

COMPETENCY ROTATION REPLICATION (exp-competency-rotation-r3)

Direct replication of original (rotation=1) with different seed. Tests seed sensitivity within same rotation condition.

PRE-REGISTERED PREDICTIONS (pooled with original + other replications):

Gamma (grounder) → source role persistence: d > 0.5 across rotations
Position effect on Alpha/Beta: significant (last speaker → sink)
Hierarchy exists (TR_range > 0.15) regardless of rotation

REPLICATION of exp-competency-rotation (N=1 pilot showed PARTIAL competency). This is N=4 total (original + 3 replications) characterization tier.

Experimental setup

Type: single_condition

Condition	Parameters
PERSONA_LIVE_ROT1	persona: PERSONA, interaction: LIVE

Factors: rotation (1)

Parameters

n_agents: 3
n_rounds: 15
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).