Eyal Ert

A Choice Prediction Competition for Market Entry Games: An Introduction

A choice prediction competition is organized that focuses on decisions from experience in market entry games (http://sites.google.com/site/gpredcomp/ and http://www.mdpi.com/si/games/predict-behavior/). The competition is based on two experiments: An estimation experiment, and a competition experiment. The two experiments use the same methods and subject pool, and examine games randomly selected from the same distribution. The current introductory paper presents the results of the estimation experiment, and clarifies the descriptive value of several baseline models. The experimental results reveal the robustness of eight behavioral tendencies that were documented in previous studies of market entry games and individual decisions from experience. The best baseline model (I-SAW) assumes reliance on small samples of experiences, and strong inertia when the recent results are not surprising. The competition experiment will be run in May 2010 (after the completion of this introduction), but they will not be revealed until September. To participate in the competition, researchers are asked to E-mail the organizers models (implemented in computer programs) that read the incentive structure as input, and derive the predicted behavior as an output. The submitted models will be ranked based on their prediction error. The winners of the competition will be invited to publish a paper that describes their model.

The Emergence of “Us and Them” in 80 Lines of Code: Modeling Group Genesis in Homogeneous Populations

Psychological explanations of group genesis often require population heterogeneity in identity or other characteristics, whether deep (e.g., religion) or superficial (e.g., eye color). We used agent-based models to explore group genesis in homogeneous populations and found robust group formation with just two basic principles: reciprocity and transitivity. These emergent groups demonstrated in-group cooperation and out-group defection, even though agents lacked common identity. Group formation increased individual payoffs, and group number and size were robust to varying levels of reciprocity and transitivity. Increasing population size increased group size more than group number, and manipulating baseline trust in a population had predictable effects on group genesis. An interactive demonstration of the parameter space and source code for implementing the model are available online.

From anomalies to forecasts: Toward a descriptive model of decisions under risk, under ambiguity, and from experience.

Experimental studies of choice behavior document distinct, and sometimes contradictory, deviations from maximization. For example, people tend to overweight rare events in 1-shot decisions under risk, and to exhibit the opposite bias when they rely on past experience. The common explanations of these results assume that the contradicting anomalies reflect situation-specific processes that involve the weighting of subjective values and the use of simple heuristics. The current article analyzes 14 choice anomalies that have been described by different models, including the Allais, St. Petersburg, and Ellsberg paradoxes, and the reflection effect. Next, it uses a choice prediction competition methodology to clarify the interaction between the different anomalies. It focuses on decisions under risk (known payoff distributions) and under ambiguity (unknown probabilities), with and without feedback concerning the outcomes of past choices. The results demonstrate that it is not necessary to assume situation-specific processes. The distinct anomalies can be captured by assuming high sensitivity to the expected return and 4 additional tendencies: pessimism, bias toward equal weighting, sensitivity to payoff sign, and an effort to minimize the probability of immediate regret. Importantly, feedback increases sensitivity to probability of regret. Simple abstractions of these assumptions, variants of the model Best Estimate and Sampling Tools (BEAST), allow surprisingly accurate ex ante predictions of behavior. Unlike the popular models, BEAST does not assume subjective weighting functions or cognitive shortcuts. Rather, it assumes the use of sampling tools and reliance on small samples, in addition to the estimation of the expected values.

Mere Position Effect in Booking Hotels Online

When travelers book hotels online, they are typically provided with a list of relevant hotels. Although presenting hotels on the screen in a list format seems appropriate for organizing the information, it creates a new (spurious) attribute for them: their position on the list. We tested experimentally whether the hotel’s position on the list affects its likelihood of being selected. Results revealed a nonlinear effect of hotel position on the list on choice: hotels that were listed at the top and bottom of the list were more likely to be chosen than those listed in the middle. This study suggests that even trivial web design choices, such as the choice of presenting data in lists, might result in nontrivial consequences on the behavior of prospective customers.

Smokers’ Decision Making: More than Mere Risk Taking

The fact that smoking is bad for people’s health has become common knowledge, yet a substantial amount of people still smoke. Previous studies that sought to better understand this phenomenon have found that smoking is associated with the tendency to take risk in other areas of life as well. The current paper explores factors that may underlie this tendency. An experimental analysis shows that smokers are more easily tempted by immediate high rewards compared to nonsmokers. Thus the salience of risky alternatives that produce large rewards most of the time can direct smokers to make bad choices even in an abstract situation such as the Iowa Gambling Task. These findings suggest that the risk taking behavior associated with smoking is not related to the mere pursuit of rewards but rather reflects a tendency to yield to immediate temptation.

A Choice Prediction Competition for Social Preferences in Simple Extensive Form Games: An Introduction

Two independent, but related, choice prediction competitions are organized that focus on behavior in simple two-person extensive form games (http://sites.google.com/site/extformpredcomp/): one focuses on predicting the choices of the first mover and the other on predicting the choices of the second mover. The competitions are based on an estimation experiment and a competition experiment. The two experiments use the same methods and subject pool, and examine games randomly selected from the same distribution. The current introductory paper presents the results of the estimation experiment, and clarifies the descriptive value of some baseline models. The best baseline model assumes that each choice is made based on one of several rules. The rules include: rational choice, level-1 reasoning, an attempt to maximize joint payoff, and an attempt to increase fairness. The probability of using the different rules is assumed to be stable over games. The estimated parameters imply that the most popular rule is rational choice; it is used in about half the cases. To participate in the competitions, researchers are asked to email the organizers models (implemented in computer programs) that read the incentive structure as input, and derive the predicted behavior as an output. The submission deadline is 1 December 2011, the results of the competition experiment will not be revealed until that date. The submitted models will be ranked based on their prediction error. The winners of the competitions will be invited to write a paper that describes their model.

Pricing decisions from experience: The roles of information-acquisition and response modes

While pricing decisions that are based on experience are quite common, e.g., setting a selling price for a used car, this type of decision has been surprisingly overlooked in psychology and decision research. Previous studies have focused on either choice decisions from experience, or pricing decisions from description. Those studies revealed that pricing involves cognitive mechanisms other than choice, while experience-based decisions involve mechanisms that differ from description-based ones. Thus, the mutual effect of pricing and experience on decision-making remains unclear. To test this effect, we experimentally compared real-money pricing decisions from experience with those from description, and with choices from experience. The results show that the mode of acquiring information affects pricing: the tendency to underprice high-probability prospects and overprice low-probability ones is diminished when pricing is based on experience rather than description. The findings further reveal attenuation of the tendency to underweight rare events, which underlies choices from experience, in pricing decisions from experience. The difference occurs because the response mode affects the search effort and decision strategy in decisions from experience.

On the Value of Experience-Based Decisions in Studying Constructs of Risk Taking

Does risk taking behavior reflect intrinsic or constructed preferences? This question is deeply rooted in psychology and decision science, but is still the subject of a lively debate. The current paper suggests that the degree of support for the idea of consistent constructs of risk taking might be contingent upon the methodology used in the experiment. Specifically, it addresses the properties of two main paradigms that have been used to study this question: decisions from description and decisions from experience (Hertwig et al., 2004). The first paradigm attends to “explicit risk taking” extracted from self-reports of risky behaviors and/or tasks in which subjects are asked to choose between descriptions of different outcomes and their associated probabilities. The second paradigm deals with “implicit risk taking.” It focuses on experience-based tasks in which people make repeated choices between options without knowing their exact payoffs and probabilities. Instead, they have to learn them from experience. Interestingly, while description-based studies typically find risk taking to be mostly situation-specific, experience-based studies show evidence of consistent individual differences, and seem able to distinguish populations associated with risk taking behaviors from controls. The current paper will first overview some evidence showing the difference between the paradigms in suggesting stable constructs of risk, and then will address properties of these paradigms that may contribute to the discrepancies.

What gets measured gets managed: A new method of measuring household food waste

The quantification of household food waste is an essential part of setting policies and waste reduction goals, but it is very difficult to estimate. Current methods include either direct measurements (physical waste surveys) or measurements based on self-reports (diaries, interviews, and questionnaires). The main limitation of the first method is that it cannot always trace the waste source, i.e., an individual household, whereas the second method lacks objectivity. This article presents a new measurement method that offers a solution to these challenges by measuring daily produced food waste at the household level. This method is based on four main principles: (1) capturing waste as it enters the stream, (2) collecting waste samples at the doorstep, (3) using the individual household as the sampling unit, and (4) collecting and sorting waste daily. We tested the feasibility of the new method with an empirical study of 192 households, measuring the actual amounts of food waste from households as well as its composition. Household food waste accounted for 45% of total waste (573 g/day per capita), of which 54% was identified as avoidable. Approximately two thirds of avoidable waste consisted of vegetables and fruit. These results are similar to previous findings from waste surveys, yet the new method showed a higher level of accuracy. The feasibility test suggests that the proposed method provides a practical tool for policy makers for setting policy based on reliable empirical data and monitoring the effectiveness of different policies over time.

Risk preferences of people with disabilities and their relation to labor market participation.

Previous studies have suggested that the participation of people with disabilities in the labor market might be affected by their risk perception, as finding a job might be perceived as an action that risks their allowance. The current study explores 2 main questions that relate to risk preferences among people with disabilities. First, it explores the potential relationship between risk preferences and employment by comparing the risk preferences of employed and unemployed people with disabilities. Second, it questions whether the risk preferences of people with disabilities are different from those of people without disabilities. To measure risk preferences in these 3 populations, we used 2 common elicitation methods: the Holt–Laury task and the balloon analogue risk task. The 2 methods complement each other, as the Holt–Laury task measures decisions from description and “explicit” risk-taking, whereas the balloon analogue risk task measures decisions from experience and “implicit” risk-taking. The results revealed no difference in risk preferences between people with and without disabilities. However, contrary to propositions from earlier studies, employed people with disabilities were found to be more risk-averse than unemployed people with disabilities. One possible interpretation of the results could be that risk aversion increases the willingness of people with disabilities to make compromises in order to participate in the labor market.