From Climate Games to Earthquakes: New Research Probes How Humans Decide Under Risk and Ambiguity

More than three centuries after Daniel Bernoulli first puzzled over a gamble that no rational person would take, the study of how human beings confront risk and uncertainty remains one of the most active frontiers in behavioral economics. A new collection of research published in Frontiers in Behavioral Economics revisits this ancient question with modern tools, spanning laboratory public goods games, household surveys in earthquake-prone regions, field experiments with farmers, and models of ethical behavior inside firms. Taken together, the four studies assembled under the research topic “Decision making under risk and ambiguity” suggest that while the mathematical machinery for modeling uncertainty has matured dramatically since 1713, the behavioral responses it seeks to explain remain stubbornly context-dependent, heterogeneous, and, in some cases, surprisingly rational.

The editors of the collection, Daniel Lee of the University of Delaware and Indu Khurana of Hampden-Sydney College, open their editorial with a observation that resonates far beyond academia: uncertainty is woven into the questions that plague individuals and societies alike. Will I have a job? Will my house withstand the next disaster? Will my children be okay? What will happen to the planet? Potential parents weigh the risk of genetic abnormalities, businesses expose themselves to possible losses, and governments assess threats ranging from armed conflict to climate crises. Crucially, the editors note, although the risks in each of these domains can be modeled by the same foundational theory, the behavioral response varies meaningfully across contexts. It is precisely this variation that the collected papers set out to illuminate.

The first study in the collection tackles one of the defining policy challenges of the century: how to persuade people to fund climate risk mitigation. Using a modified public goods game, the authors of “Risk and ambiguity” explore preferences for mitigation strategies, as opposed to abatement strategies, in combating climate risk. The experimental design distinguishes between group and individual decision making and, importantly, accounts for heterogeneous subject types. The central finding is subtle but consequential: introducing an ambiguous information environment has no effect on average contributions to mitigation, yet it significantly inhibits mitigation contributions among ambiguity-averse subjects. In other words, the population-level average conceals a meaningful behavioral divide that only becomes visible when subject heterogeneity is taken seriously.

Perhaps the most striking result from this climate game concerns what people are willing to pay to escape ambiguity. More than half the time, participants wished to have access to information that would resolve the ambiguous environment, and they actually used that information once they obtained it. This willingness to pay for information carries direct policy implications. If ambiguity itself suppresses contributions to public mitigation efforts, then investments in clearer communication about climate risks, better probabilistic forecasts, and more transparent data could pay for themselves by unlocking cooperative behavior that uncertainty currently suppresses.

A second paper, “Beyond average preparedness,” brings the analysis of ambiguity into the seismic zone. Author Nkouaga uses online polling data to explore earthquake preparedness among homeowners in the New Madrid Seismic Zone, the intraplate fault system responsible for some of the largest earthquakes in United States history. Much like the ambiguity confronting climate game participants, earthquakes cannot be predicted, a point the United States Geological Survey emphasizes in its public guidance. Against this backdrop of irreducible uncertainty, Nkouaga finds that what the study terms “Community Engagement,” including discussions with family and friends, participation in community events, and interactions with neighbors regarding earthquake hazards, is the strongest predictor of earthquake preparedness. The finding suggests that preparedness is not merely a function of individual risk perception but of social embeddedness.

The methodological rigor of the preparedness study deserves particular attention. Nkouaga’s results hold under both ordinary least squares and quantile regression approaches at all quantiles, indicating that the main findings are not driven by an extreme portion of the preparedness distribution. Yet the quantile analysis also reveals something the average would hide: variables such as household income have important threshold values, meaning their effect on preparedness is not uniform across the distribution. This nuance connects back to the information-purchasing behavior observed in the climate game. If income thresholds determine who can act on risk information, then the costs of revealing information may fall unevenly across a population, and policies that subsidize information access could have disproportionate effects at specific points in the income distribution.

The third paper, “Exposure to climate-change,” asks whether long-term exposure to climate-change-related extreme weather events is associated with farmers’ risk preferences. Using a mixed-methods design that combines an artefactual field experiment with a household survey, the authors find that farmers in risky locations are less risk-averse. At first blush, the result may seem obvious: people who live with frequent hazards should get used to them. But the editors argue the finding is deeply profound, connecting to a long line of literature on rational decision-making in unique environments. Farmers who repeatedly experience extreme weather may rationally update their beliefs about the true distribution of risks, or they may sort into risky locations precisely because they tolerate risk. Either interpretation implies that risk preferences are not fixed traits but responses to lived experience, a conclusion with major implications for how researchers should elicit and interpret preference parameters.

The study’s use of a non-WEIRD sample, meaning participants drawn from outside the Western, Educated, Industrialized, Rich, and Democratic populations that dominate experimental economics, also lends credibility to ongoing debates about the external validity of laboratory findings. Understanding behavioral responses to climate events in the developing world, the editors note, can help policymakers design better solutions for mitigation and abatement, such as index insurance, a financial instrument that pays out based on observable triggers like rainfall rather than individually verified losses. Recent work on the welfare consequences of compound risks in index insurance underscores that the design of such instruments depends critically on how beneficiaries perceive and respond to layered uncertainty.

The fourth paper shifts the setting from fields and fault lines to the office. “Internal security” examines decisions in an intra-office setting, presenting a fully mediated model of the impacts of employability on unethical behavior undertaken in favor of a home company’s bottom line. The paper speaks to the risk inherent in a real options framework when employees and firms consider exits and next steps. In plain terms, workers who perceive their outside employment options as secure may behave differently, and potentially less ethically, toward their current employer than those who feel trapped. The result extends the study of decision under uncertainty from markets and disasters into the microfoundations of organizational behavior, where the stakes are reputational and legal rather than seismic.

Across all four papers, the editors identify methodological themes that will be familiar to behavioral scientists. The climate game highlights the importance of contextual, as opposed to neutral, framing, echoing calls for experimentalists to think deeply about which parameters they are actually eliciting. The farmers study blends methods and samples in ways that speak directly to validity debates. Yet the editors are careful to emphasize that there is still more work to do. Future research, they argue, will be more interdisciplinary, tying together frameworks from multiple fields. And the collected papers point toward an even more insidious conclusion: given incomplete data and opaque, complex models, we are constantly exposing ourselves to new forms of risk and uncertainty. As human interactions with artificial intelligence, machine learning, and algorithmic interfaces proliferate, a more thoroughly developed understanding of how humans approach risk and ambiguity will only become more essential for navigating the uncertainties of the digital age, and whatever lies beyond.

Subject of Research: Behavioral economics of decision making under risk and ambiguity

Article Title: Editorial: Decision making under risk and ambiguity

Article References: Lee, D., & Khurana, I. (2026). Editorial: Decision making under risk and ambiguity. Frontiers in Behavioral Economics, 5, Article 1960515. https://doi.org/10.3389/frbhe.2026.1960515

Image Credits: AI Generated

DOI: 10.3389/frbhe.2026.1960515

Keywords: risk, ambiguity, behavioral economics, decision making, climate risk, earthquake preparedness, public goods game, risk preferences, index insurance, uncertainty, experimental economics, real options