Thesis Scope and Experimental Baseline

Research question: How do different voting rules influence the temporal evolution of participation rates and inequality in a simple multi-agent system with adaptive agents?

Related documents:

• docs/research/thesis_measurement_spec.md
• docs/research/metric_glossary.md
• docs/research/thesis_model_concepts.md

1. Scientific Scope

The thesis studies a fixed simulation environment where voting rule is the primary intentionally varied factor in rule-comparison runs.

Participation and altruism are adaptive, but their update mechanisms are fixed within a given experiment set.

2. Baseline Model Settings

Baseline family:

• quality_target_mode = puzzle
• participation_signal_mode = group_relative_delta_rel_party

• altruism_mode = satisfaction

• satisfaction-response parameters:

  • altruism_satisfaction_theta
  • altruism_satisfaction_slope
  • altruism_response_gamma

Implemented voting-rule mapping (rule_idx):

• 0 = plurality
• 1 = approval
• 2 = utilitarian
• 3 = borda
• 4 = schulze
• 5 = random

3. Rule-Family Framing for Thesis Inference

• confirmatory canonical family: utilitarian (2), borda (3), schulze (4)
• reference family: plurality (0), random (5)
• context-only calibration arm: approval (1)

Family labels define reporting roles, not implementation differences in runtime mechanics.

4. Primary Outcomes

Primary outcome families are evaluated as time-series and run-level summaries:

• participation dynamics (turnout)
• resource inequality (gini_assets from steps.gini_index)
• experiential inequality (gini_dissatisfaction from agents.dissatisfaction_value)
• outcome-quality trajectory (quality_distance aggregated from area_steps with mode-aware source)

Interpretation note:

• assets denotes simulation resource/capacity state, not necessarily currency.

5. Secondary Descriptive Analyses

Secondary (non-confirmatory) views include:

• group-level turnout and participation-composition dynamics
• participation/inequality co-movement patterns
• benchmark-reference distance diagnostics from logged artifacts

6. Out of Scope for Confirmatory Claims

• strategic voting equilibria
• empirical calibration against real election datasets
• normative policy prescriptions
• claims of globally optimal democratic design