Behavioral Sensitivity Modeling in Financial Decision Systems: Integrating Investor Psychology with Predictive Analytics

Abstract

The research examines the interaction between investor psychology and algorithmic decision-support systems in current financial contexts. It examines how cognitive biases, emotional responses, and reliance on predictive analytics all influence investor behavior in uncertain conditions. Employing a qualitative methodology based on thematic and grounded theory analysis, data was gathered from 200 participants, comprising retail investors, financial analysts, and portfolio managers, to discern recurring behavioral patterns and psychological responses to AI-driven financial tools. The results indicate three interconnected dimensions: Decision Sensitivity under Uncertainty, Overconfidence Bias, and Trust in Predictive Analytics that collectively establish the basis of the proposed Behavioral Sensitivity Model. Findings demonstrate that uncertainty enhances emotional responses and algorithm-based choices; overconfidence encourages selective dependence on algorithmic results, resulting in a new distortion known as algorithmic validation bias; conversely, confidence in transparent, explainable AI systems mitigates these effects, facilitating more consistent and rational decision-making. The work conceptually contributes by defining behavioral sensitivity as a multidimensional construct that connects cognitive bias, emotional elasticity, and human–AI interaction, while also enhancing behavioral finance through the concept of dynamic bias adaptation in technology-mediated environments.