Neuroeconomics aims to discover mechanisms of economic decision, and express them mathematically, to predict observed choice. While the contents of neuroeconomic models and evidence are obviously different than in traditional economics, (some of the) goals are identical: to explain and predict choice, the effects of comparative statics, and perhaps make interesting new welfare judgments that are defensible. To this end, Paul Glimcher's important book carefully describes how economics, psychological, and neural levels of explanation can be linked (a structure which has been successful in visual neuroscience). As Glimcher shows, the neural evidence is quite strong for a process of learning valuations through prediction error, and a simple model of neural valuation and comparison that corresponds to random utility (though subject to normalization, which produces menu effects). There is also rapidly growing evidence for more complicated constructs in behavioral economics, including prospect theory's account of risky choice, hyperbolic time discounting, level-k models of games, and social preferences corresponding to internal reward based on what happens to other agents.

Microeconomic theory typically concerns exchange between individuals or firms in a market setting. To make predictions precise, individuals are usually assumed to use the laws of probability in structuring and revising beliefs about uncertainties. Recent evidence, mostly gathered by psychologists, suggests probability theories might be inadequate descriptive models of individual choice. (See the books edited by Daniel Kahneman et al., 1982a, and by Hal Arkes and Kenneth Hammond, 1986.)

To test whether naturally occurring markets can be strategically manipulated, $500 and $1,000 bets were made, then canceled, at horse racing tracks. The net effects of these costless temporary bets give clues about how market participants react to information large bets might contain. The bets moved odds on horses visibly (compared to matched‐pair control horses with similar prebet odds) and had a slight tendency to draw money toward the horse that was temporarily bet, but the net effect was close to zero and statistically insignificant. the results suggest that some bettors inferred information from bets and others did not, and their reactions roughly canceled out.

Bottom-up stimulus-driven visual salience is largely automatic, effortless, and independent of a person’s “top-down” perceptual goals; it depends only on features of a visual stimulus. Algorithms have been carefully trained to predict stimulus-driven salience values for each pixel in any image. The economic question we address is whether these salience values help explain economic decisions. Our first experimental analysis shows that when people pick between sets of fruits that have artificially induced value, predicted salience (which is uncorrelated with value by design) leads to mistakes. Our second analysis uses evidence from games in which choices are locations in images. When players are trying to cooperatively match locations, predicted salience is highly correlated with the success of matching (r = .57). In competitive hider-seeker location games, players choose salient locations more often than predicted by the unique Nash equilibrium. This tendency creates a disequilibrium “seeker’s advantage” (seekers win more often than predicted in equilibrium). The result can be explained by level-k models in which predicted stimulus-driven salience influences level-0 choices and thereby influences overall perceptions, beliefs, and choices of higher-level players. The third analysis shows that there is an effect of visual salience in matrix games, but it is small and statistically weak. Applications to behavioral IO, price and tax salience, nudges and design, and visually influenced beliefs are suggested.

Many alternative theories have been proposed to explain violations of expected utility (EU) theory observed in experiments. Several recent studies test some of these alternative theories against each other. Formal tests used to judge the theories usually count the number of responses consistent with the theory, ignoring systematic variation in responses that are inconsistent. We develop a maximum-likelihood estimation method which uses all the information in the data, creates test statistics that can be aggregated across studies, and enables one to judge the predictive utility-the fit and parsimony-of utility theories. Analyses of 23 data sets, using several thousand choices, suggest a menu of theories which sacrifice the least parsimony for the biggest improvement in fit. The menu is: mixed fanning, prospect theory, EU, and expected value. Which theories are best is highly sensitive to whether gambles in a pair have the same support (EU fits better) or not (EU fits poorly). Our method may have application to other domains in which various theories predict different subsets of choices (e.g., refinements of Nash equilibrium in noncooperative games).

This paper considers the measurement of consumer loss aversion in product markets. We introduce a test based on a “substitution effect,” focusing on how the end of a sale affects sales not of the good itself, but a substitute good. Such an effect cannot be easily confounded with consumer stockpiling. Using a unique dataset from an online hardware retailer, we find evidence consistent with consumer loss aversion. Moreover, we find that less experienced consumers suffer a more prominent loss aversion bias compared to more experienced consumers.

This paper tests two explanations for apparent undersaving in life cycle models: bounded rationality and a preference for immediacy. Each was addressed in a separate experimental study. In the first study, subjects saved too little initially—providing evidence for bounded rationality—but learned to save optimally within four repeated life cycles. In the second study, thirsty subjects who consume beverage sips immediately, rather than with a delay, show greater relative overspending, consistent with quasi-hyperbolic discounting models. The parameter estimates of overspending obtained from the second study, but not the first, are in range of several empirical studies of saving (with an estimated β = 0.6–0.7).

We conducted experiments in Vietnamese villages to determine the predictors of risk and time preferences. In villages with higher mean income, people are less loss-averse and more patient. Household income is correlated with patience but not with risk. We expand measurements of risk and time preferences beyond expected utility and exponential discounting, replacing those models with prospect theory and a three-parameter hyperbolic discounting model. Comparable risk parameter estimates have been found for Chinese farmers, using our method. (C83, D12, O12, P38)

We test for consumption smoothing using bankruptcy data on players in the National Football League (NFL), who typically earn several million dollars during an income spike that lasts a few years. The life-cycle hypothesis predicts that players should save substantially while playing and then have little risk of bankruptcy post-NFL. However, players in our sample begin to file for bankruptcy soon after they stop playing and continue filing at a high rate through at least the first 12 years of retirement. Players' total earnings and career lengths have surprisingly little effect on the risk of bankruptcy.

We report experiments on sender-receiver games with an incentive for senders to exaggerate. Subjects “overcommunicate”—messages are more informative of the true state than they should be, in equilibrium. Eyetracking shows that senders look at payoffs in a way that is consistent with a level-k model. A combination of sender messages and lookup patterns predicts the true state about twice as often as predicted by equilibrium. Using these measures to infer the state would enable receiver subjects to hypothetically earn 16–21 percent more than they actually do, an economic value of 60 percent of the maximum increment.