Trust-Based Adaptation in Human Decision Systems: Introducing the Socio-Evolutionary Algorithm (SEA)_PreSubmission

Human behaviour rarely unfolds in isolation but adapts dynamically through trust, imitation, and exploratory trial-and-error within social and institutional contexts. Existing computational approaches capture only fragments of this process. The Socio-Evolutionary Algorithm (SEA) integrates credibility-based trust, bounded imitation, and exploratory variation into a generalisable framework for socially adaptive decision-making, evaluated across economic and social domains. We assess SEA in ten heterogeneous cross-sectional populations and two longitudinal datasets (25-wave Youth Panel; 14-wave YP2007), using both actual and statistically augmented data with demonstrated equivalence. SEA reproduces key adaptation dynamics and suggests improved fit over Bayesian, hybrid, and conventional evolutionary benchmarks. An enhanced variant with adaptive trust learning (αₜ) and uncertainty-sensitive mutation (μₜ) indicates further robustness. SEA offers a transparent and reproducible approach to modelling socially embedded adaptation while defining its generality within validated economic and social contexts.