The Effect of Personalization Algorithm-Based Dark Patterns on Consumer Decision Distortion: Integration of Business Ethics and Behavioral Economics Perspectives on E-Commerce Platforms

Abstract

The rapid advancement of artificial intelligence (AI)-driven personalization in e-commerce has introduced algorithm-based dark patterns that subtly manipulate consumer behavior and potentially distort decision-making processes. This study aims to analyze the effect of personalization-based dark patterns on consumer decision distortion by integrating perspectives from behavioral economics and business ethics. A quantitative approach was employed using a scenario-based survey involving 212 active e-commerce users. Data were analyzed using Partial Least Squares Structural Equation Modeling (PLS-SEM) to examine both direct and indirect relationships among variables. The results indicate that personalization-based dark patterns have a positive and significant effect on consumer decision distortion. Additionally, cognitive biases, particularly loss aversion and anchoring effect, significantly mediate this relationship. Loss aversion is found to be the more dominant mediating factor, suggesting that consumers are more influenced by the fear of missing out than by reference-based judgments. These findings confirm that digital consumers are not fully rational, as their decisions are shaped by psychological biases exploited through system design. In conclusion, algorithm-based dark patterns significantly contribute to distorted consumer decisions through both direct influence and cognitive mechanisms. This study recommends the implementation of more transparent and ethical interface designs, as well as stronger regulatory frameworks to limit manipulative practices. The findings provide theoretical implications by integrating technological, behavioral, and ethical perspectives into a unified framework, and practical implications for developers and policymakers in promoting responsible and sustainable digital platforms.