Benson Tsz Kin LEUNG

Benson Tsz Kin LEUNG

Assistant Professor in Economics

Research

You can also find my articles on my Google Scholar profile.

Published Articles

1. Memory Overload and Asymmetric Under-reaction, 

Previously circulated as “Information Overload and Confirmation Bias”. With Lorenz Götte and Hua-Jing Han. Accepted at Games and Economic Behavior, July 2026.

Abstract We show that stronger memory overload results in a disproportionately greater under-reaction to belief-challenging information compared to belief-confirming information. In an experiment, we keep constant the information received by subjects but vary the difficulty of memorizing the signals. In the treatment condition, the preceding signal disappears as the next signal appears. In the control condition, the preceding signal remains visible. There is stronger memory overload in the treatment than in the control condition. We find that in the treatment condition, subjects under-react when they receive belief-challenging information but not when they receive belief-confirming information. Further analysis supports the mechanism in which it is more cognitively taxing to memorize belief-confirming than belief-challenging signals. However, it does not support the mechanism in which subjects have a fixed amount of cognitive resources to allocate to different signals. Our results caution against information campaigns that ignore the role of cognitive load and attention.

Paper | Experimental Instruction

2. Innovation Diffusion among Case-based Decision-makers, 

Journal of Economics & Management Strategy 35, no. 2 (2026): 311-325.

Abstract This paper analyzes a model of innovation diffusion with case-based individuals a la Gilboa and Schmeidler (1995, 1996, 1997), who decide whether to consume an incumbent or a new product based on their and their social neighbors' previous consumption experiences. I analyze how diffusion pattern changes with individual characteristics, innovation characteristics and social network. In particular, radical innovation leads to higher initial speed but lower acceleration compared to increment innovation. Social network with stronger overall social tie, lower degree of homophily or higher exposure of reviews from early adopters speed up diffusion of innovation.

Paper | Journal | arXiv

3. Risk Attitude and Belief Updating: Theory and Experiment, 

With Evelyn Y.H. Huang. Frontiers in Psychology 14 (2023): 1281296. Abstract published in CogSci 2023.

Abstract Despite the importance of risk attitude in decision-making, its role in belief updating has been overlooked. Using economic theory, we analyze a dual-self equilibrium where an individual first updates her belief about an uncertain state, then takes an action to maximize her payoff. We show that stronger risk aversion drives more conservative actions, and thus decreases the instrumental value of information relative to the importance of belief-based utility. As a result, the relationship between risk attitude and belief updating depends on the nature of the belief-based utility. With self-relevant information, stronger risk aversion leads to more belief change; whereas with self-irrelevant information, stronger risk aversion leads to less belief change. Our experimental results concur with the theoretical predictions with two settings where subjects update their belief about their IQ and a randomly drawn number, respectively. We discuss implications on persuasion, advertisements, and political campaigns.

Paper

4. Limited Cognitive Ability and Selective Information Processing, 

Games and Economic Behavior 120 (2020): 345-369.

Abstract This paper studies the information processing behavior of a decision maker (DM) who can only process a subset of all information he receives: before taking an action, the DM receives sequentially a number of signals and decides whether to process or ignore each of them as it is received. The model generates an information processing behavior consistent with that documented in the psychological literature: first, the DM chooses to process signals that are strong; second, his processing strategy exhibits confirmation bias if he has a strong prior belief; third, he tends to process signals that suggest favorable outcomes (wishful thinking). As an application I analyze how the Internet and the induced change in information availability affects the processing behavior of the DM. I show that providing more/better information to the DM could strengthen his confirming bias.

Paper | Online Appendix

Working Paper

1. Distracted from Comparison: Product Design and Advertisement with Limited Attention, 

With Johannes Johnen. Revised and Resubmitted (3rd Round) at AEJ-Micro, June 2026.

Abstract We study how firms choose designs—of their products or product information—to divert consumers' limited attention away from price comparison or towards it. Firms choose designs to affect the dispersion of product match values and thereby how consumers allocate their limited attention. Consumers with limited attention trade off breadth and depths of search: they either study fewer products in detail to learn their match value, or superficially browse and compare prices of more products. We highlight a novel distraction effect of designs. Firms combine larger prices with designs that disperse match values to distract consumers from price-comparison. We show that more-detailed product information disperse match values and allows firms to distract consumers more effectively from price comparison. This way, interventions that allow firms to disclose more-detailed product information weaken competition and decrease consumer surplus. In turn, interventions that make information coarser and more easily-available information—like energy-efficiency labels and front-package food labels like nutriscores—increases competition and consumer surplus. These findings connect evidence of various informational interventions in the context of pension funds, advertisements, and food labels.

Paper | CEPR

2. Learning in a Small/Big World, 

Revised and Resubmitted at IER, May 2026.

Abstract Complexity and limited ability affect how we learn and make decisions under uncertainty. Using finite automata to model belief formation, this paper studies the characteristics of optimal learning behavior in small and big worlds, where the complexity of the environment is low and high, respectively, relative to the cognitive ability of the decision maker. The optimal long-run average payoff and distribution of actions are well approximated by the Bayesian benchmark in very small worlds, but the gap widens as the worlds get bigger. In big worlds, it can be optimal for the decision maker to ignore some states entirely, concentrating limited cognitive resources on a subset of states. This ``optimal ignorance'' is reminiscent of well-documented non-Bayesian learning patterns such as the use of heuristics, correlation neglect, and inattentive learning. The results establish a relationship between complexity, cognitive ability, and the scope for non-Bayesian learning behavior.

Paper | arXiv | Online Appendix

3. Dancing Around The Elephant In The Room: Information Avoidance and Willful Justification, 

With Evelyn Y.H. Huang. August 2025 (Draft Available Upon Request).

Abstract Does exposing information avoiders to information induce belief and behavioral change? In an experiment, subjects make two investment decisions as they choose between two bankers. In between, they decide whether to receive or avoid a piece of free information about whether their first decision is good or bad. Crucially, we randomize information provision such that information avoiders may receive information, and non-avoiders may receive no information. First, subjects avoid information when they are pessimistic about their first decisions, implying that subjects avoid due to cognitive dissonance. Second, we find that upon receiving the same piece of positive information, avoiders and non-avoiders behave similarly; upon receiving no information, compared to non-avoiders, avoiders are more optimistic about and more likely to persist with their first decisions; upon receiving negative information, avoiders update their belief on their first decisions similarly as non-avoiders, interestingly they are more likely to persist with the first decisions and justify by updating negatively about the other banker. The results suggest a limited impact of information campaigns that "force" individuals to receive and pay attention to information.
4. A Simple Model of Memory-Based Belief Formation, 

September 2023. [Working on major changes!]

Abstract This paper presents a simple model of memory-based belief formation. An individual has a memory of signals, and upon receiving a prompt, recalls signals from his memory and forms belief using a naïve Bayes' formula. The recall probability of each signal depends on the cues of the signals and the prompt. The recall probability function can capture different stylized results of memory research, including how similarity, interference, salience, previous retrieval, etc. affect memory retrieval. I analyze three adaptations of the model and show that, first, varying recall probabilities w.r.t. likelihood ratio can lead to asymptotic erroneous belief. Second, the persuader prefers to send a less informative but more associative signal when the receiver has a large memory. Third, individuals overreact to new signals when they have a large memory, and will strengthen their strong prior beliefs even after receiving a belief-challenging signal.

Paper

5. Social Media & Political Competition: When lowering entry barriers strengthens incumbency advantage, 

With Pinar Yildirim. November 2020.

Abstract An increasing number of politicians are relying on cheaper, easier to access technologies such as online social media platforms to communicate with their constituency. These platforms present a cheap and low-barrier channel of communication to politicians, potentially intensifying political competition by allowing many to enter political arena. In this study, we demonstrate that lowering costs of communication, which allows many entrants to come into a competitive market, can strengthen an incumbent's position when the newcomers compete by providing more information to the voters. We show an asymmetric bad-news-good-news effect where early negative news hurts the challengers more than the positive news benefit them, such that in aggregate, an incumbent politician's chances of winning is higher with more entrants in the market. Our findings indicate that communication through social media and other cheap platforms can intensify competition, however incumbency advantage may be strengthened rather than weakened as an outcome of higher number of entrants into the political market.

Paper

Work in Progress

1. The Determinants of the Demand of Information, 

With Leifu Zhang.

Abstract Coming soon!

Dormant Projects

1. Information Disclosure in a Duopoly, 

April 2018.

Abstract This paper studies firms' disclosure decisions of product information and the welfare implication of mandatory information disclosure. I show that the incentive to (not) disclose product information could be summarized in two channels: the effect on the relative market power over rivals and the competitiveness in the market. Although mandatory information disclosure allows consumers to pick a better match, it could increase the average price in the market and decrease consumer welfare. I then present simple conditions on when mandatory disclosure always improve consumers welfare, and when it might hurt consumers.

Paper