Seeing Through Risk: A Symbolic Approximation of Prospect Theory

2025-04-20

AI Summary

This paper addresses the longstanding trade-off between interpretability and behavioral fidelity in risk-decision modeling by introducing the first symbolically grounded, interpretable prospect theory model. Methodologically, it replaces black-box utility and probability weighting functions with explicit symbolic representations of core psychological mechanisms—such as loss aversion and framing effects—formalized as effect-size-driven symbolic features; these are then mapped to interpretable, psychologically meaningful parameters via symbolic regression and mathematical formalization. Contributions include: (1) achieving predictive performance on par with state-of-the-art black-box models while preserving strict adherence to behavioral theory; (2) accurately reproducing canonical risk-preference phenomena (e.g., reflection effect, certainty effect) on synthetic data; and (3) ensuring all model parameters possess unambiguous, theory-grounded psychological interpretations—thereby bridging the gap between behavioral decision theory and interpretable machine learning.

Technology Category

• Reasoning under Uncertainty: Decision/Utility Theory
• Cognitive Modeling & Cognitive Systems: Symbolic Representations
• Knowledge Representation and Reasoning: Preferences

Application Category

• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

We propose a novel symbolic modeling framework for decision-making under risk that merges interpretability with the core insights of Prospect Theory. Our approach replaces opaque utility curves and probability weighting functions with transparent, effect-size-guided features. We mathematically formalize the method, demonstrate its ability to replicate well-known framing and loss-aversion phenomena, and provide an end-to-end empirical validation on synthetic datasets. The resulting model achieves competitive predictive performance while yielding clear coefficients mapped onto psychological constructs, making it suitable for applications ranging from AI safety to economic policy analysis.

Problem

• Symbolic modeling for interpretable risk decision-making
• Replacing opaque Prospect Theory elements with transparent features
• Validating model on framing, loss-aversion, and synthetic datasets

Innovation

• Symbolic modeling framework for risk decisions
• Transparent features replace utility curves
• Mathematical formalization with empirical validation