Abstract: As e-commerce platforms increasingly deploy algorithms to implement price discrimination, consumers who detect such practices often retaliate against the offending platform, yet whether and how this retaliation propagates beyond a single platform remains poorly understood. Prior research has documented diverse forms of algorithmic price discrimination and the algorithmic biases consumers hold toward them, such as algorithm aversion and algorithmic transference, but has rarely asked whether and how these distinct forms of discrimination give rise to a cascading retaliation effect that spreads across platforms and contexts, or under what individual and situational conditions this spillover intensifies. Equally unclear is how classic cognitive biases shape the strength of this contagion. To address the theoretical gap, this study investigates these chain-reaction behavioral patterns and establishes a comprehensive analytical framework through two interconnected studies.
Study 1 aims to develop a multidimensional classification system for algorithmic price discrimination intensity, categorizing it into four levels based on the degree of discrimination: dynamic pricing, consumer segment-based pricing, personalized pricing, and personalized dynamic pricing. The study proposes and confirms that higher algorithmic price discrimination intensity systematically undermines three dimensions of consumer-perceived fairness: distributive fairness (triggered by the magnitude of price disparities), procedural fairness (stemming from opaque pricing mechanisms), and interactional fairness (triggered by the algorithm's lack of emotion and human-centric considerations). These three dimensions collectively mediate the transmission path from algorithmic price discrimination to retaliatory chain reactions. Furthermore, the study identifies two key boundary conditions. Price sensitivity amplifies the sense of distributive unfairness, thereby intensifying the retaliatory chain reaction. Conversely, algorithmic explainability acts as a psychological buffer; by enabling consumers to understand algorithmic rules and recognize differences in algorithmic logic, it disrupts the perception of algorithmic homogeneity, effectively confining consumer dissatisfaction to the current platform.
Drawing on the perspective of bounded rationality, Study 2 aims to reveal the cognitive mechanisms driving this cross-platform retaliation contagion, identifying the interplay between fairness perceptions and two cognitive biases: loss aversion and the availability heuristic. First, loss aversion, the phenomenon where the psychological pain of a loss outweighs the pleasure of an equivalent gain, systematically amplifies the sense of distributive unfairness. Consumers with higher levels of loss aversion perceive greater unfairness and exhibit more intense retaliatory chain reactions. Second, the availability heuristic influences the cognitive salience of the event. Variations, such as the timing of algorithmic price discrimination incidents and the intensity of the negative emotions elicited, lead to differences in how easily consumers recall these events, which in turn influence their judgments regarding the likelihood of platform discrimination. The discrimination-related memory is activated by perceptions of unfairness, subsequently determining the intensity of the resulting spillover effects.
This study makes four major theoretical contributions. First, it constructs an algorithmic price discrimination classification system from the consumer’s perspective, extending classical price discrimination theory to the context of algorithm-driven markets. Second, it examines the three-dimensional theory of fairness within the context of AI interfaces, offering new insights into the understanding of algorithmic bias. Third, it shifts the paradigm of consumer retaliation from isolated "stimulus-response" frictions to systemic, cross-platform, and cross-contextual boycotts, thereby providing a novel explanation for the spread of distrust within digital ecosystems. Fourth, it expands the scope of the "algorithmic spillover effect" from technical malfunctions to the realm of pricing, revealing how bounded rationality can trigger widespread algorithmic distrust.
By mapping the complete behavioral trajectory, ranging from types of discrimination and perceptions of three-dimensional fairness to cognitive biases and chain reactions of retaliation, this study redefines algorithmic pricing discrimination as a systemic risk at the ecosystem level rather than a localized issue confined to a single firm, thereby providing a crucial behavioral basis for platform governance in the AI ​​era.

Key words: consumer behavior, algorithmic price discrimination, psychological mechanism, cognitive bias, spillover effect