EXP-OPPONENT

informative

Opponent Processing — Competitive vs Cooperative Persona Framing

2026-03-09 L2 level paper3 1 runs $0.00

new

premortem

mock

real

metrics

analyze

review

krit

done

In plain language

What shapes hierarchy more — cooperation or competition? Three AI agents with adversarial styles (critic, contrarian, skeptic) interact in a shared text space. We compare with the same agents but with complementary styles (analyst, associator, practitioner).

What we found

No predictions scored

Predictions we made before running

0/4 confirmed

▶ Technical details

Research hypothesis

Competitive/adversarial persona framing produces different trophic hierarchy dynamics than complementary/cooperative framing. Two competing predictions: (A) Adversarial roles create STRONGER differentiation — distinct vocabulary domains for attack vs defense generate directional flow (critic originates critical vocabulary, target adopts it to respond). (B) Adversarial mirroring COLLAPSES hierarchy — opponents converge on shared vocabulary as they argue about the same claims, destroying the export/import gradient that creates trophic structure. Baseline comparison: exp-asym PERSONA_LIVE (cooperative, N=3, 9 runs, TR range 0.444).

Experimental setup

Type: factorial

Condition	Parameters
ADVERSARIAL_LIVE	persona: ADVERSARIAL, interaction: LIVE
ADVERSARIAL_ISOLATED	persona: ADVERSARIAL, interaction: ISOLATED

Factors: persona (ADVERSARIAL) × interaction (LIVE, ISOLATED)

Parameters

n_agents: 3
n_rounds: 48
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).