Enhancing Investor Rationality: A Review of Behavioural Biases and the De-Biasing Potential of Artificial Intelligence in Investment Decision Making

Abstract

Traditional financial theories assume that investors are rational individuals who make decisions based on all the information available at their disposal. However, insights from behavioural finance research suggest that investors frequently deviate from rationality because of the psychological and cognitive biases. These biases affect judgment and can result in suboptimal investment outcomes. The purpose of this study is to present a narrative review of important behavioural biases that influence investment decisions, as well as to examine how artificial intelligence-driven tools help in minimising their impact. The study adopts a narrative review approach, drawing insights from both foundational and recent literature on behavioural finance and financial technology. It analyses prior research to understand how different biases shape investor behaviour and decision-making processes. The review highlights cognitive biases, namely confirmation bias, anchoring bias, and overconfidence bias, that significantly influence investment choices. In addition, emotional biases further distort the decision-making process. The findings also indicate that AI-driven tools can support investors by offering objective, data-driven information, personalised recommendations, and help minimise the influence of biases, thereby improving decision quality. Overall, this study contributes to the growing literature by offering a clear understanding of the role of behavioural biases in investment decision-making and the corrective potential of AI-driven tools to address them. However, this study is limited by its reliance on secondary literature and the absence of empirical validation, which provides scope for future empirical research. It provides practical insights for investors, practitioners, and researchers seeking to navigate financial markets more effectively in a technology-driven environment.