Back to search
2108.00966

Rationality and Reciprocity of Opinion Dynamics in Games

Shinkyu Park, Anastasia Bizyaeva, Mari Kawakatsu, Alessio Franci, Naomi Ehrich Leonard

correctmedium confidence
Category
Not specified
Journal tier
Strong Field
Processed
Sep 28, 2025, 12:56 AM

Audit review

The paper proves a Z2-symmetric pitchfork in the 2-agent, 2-strategy opinion dynamics and derives the Lyapunov–Schmidt (LS) reduced normal form with explicit parameter dependence; see Theorem 1 and its reduction dzc/dτ = [(η^{-1}p + (α+γ)ũ)/(2(α^2−αγ+γ^2))]zc − zc^3 + [(p+2p⊥)/(2(α^2−αγ+γ^2))] + h.o.t. (their Eq. (14), Appendix Eqs. (18)–(19)) . The candidate solution independently reconstructs the same normal form by restricting to the symmetric invariant subspace, expanding tanh to cubic order, and identifying the unfolding directions “tilt” (∝ p zc) and “bias” (∝ p+2p⊥). Small discrepancies appear only in intermediate coefficients (a factor-of-two in the p zc slope and the omission of d in the time rescaling), but after the standard normal-form/time rescalings the candidate’s final equation coincides with the paper’s stated normal form and parameter dependence. The paper’s derivation via LS partial derivatives (Appendix: Jacobian singularity at u=2/(α+γ), cubic coefficient −2(α^2−αγ+γ^2), and the mixed p, p⊥ derivatives) is correct and complete , and the model solution reaches the same conclusion by a different (but consistent) route.

Referee report (LaTeX)

\textbf{Recommendation:} minor revisions

\textbf{Journal Tier:} strong field

\textbf{Justification:}

The manuscript cleanly formulates and analyzes a model that blends rationality and reciprocity, and it provides a rigorous, interpretable bifurcation mechanism for cooperation in the prisoner’s dilemma. The LS reduction and parameter dependence of the unfolding are correctly and transparently derived. Minor inconsistencies in time normalization and a few derivational details can be polished easily; otherwise the work is technically sound and clearly presented.