When I came to Decision Research to work with Paul Slovic and Sarah Lichtenstein, Paul was already working with Howard Kunreuther on a groundbreaking study of flood and earthquake insurance purchasing behavior. For this reason, I can say that Howard has been a part of my professional life since I left graduate school.

It took me a while to fully grasp what he and the Disaster Insurance Project were doing. For someone like me, trained in cognitive experimental psychology and mathematics, it was a brave new world: collaborating with economists, conducting population-based surveys, studying complex, ill-conceived real-world decisions rather than stylized experimental decisions designed to delineate theoretical approaches, and trying to provide information for complicated public policies like the National Flood Insurance Program.

There were conceptual precursors to the project. Ward Edwards, the founder of decision science, distinguished between expected value, expected utility, and subjective expected utility decision models, depending on whether people relied on their own judgment (utilities, subjective probabilities) or an external standard (monetary values, frequentist probability estimates). Previous experiments had compared these implications in tightly controlled settings. But Howard and his colleagues surveyed homeowners about actual, fateful decisions.

I had a cameo appearance on the project, designing some stylized experiments and assisting our programmer Bernie Corrigan in developing an interactive computer game that simulated running a farm, including insurance decisions. Thanks to the generosity of Howard, Paul, and others, I was close enough to the action to learn some lessons that have stayed with me throughout my career and were the cornerstones of Howard’s life’s work.

Disciplined empathy is at the heart of all decision research. We need to listen to people to understand and respond to their concerns. We need to care about them enough to put those concerns above our professional interests, such as testing or promoting preferred theories.

Details are important, especially when it comes to questions about important decisions. We need to understand the technical details of decisions well enough to ask about the choices people are facing. These details include the likelihood and severity of the disaster, insurance coverage, and other details such as deductibles and premiums. We also need to be sure that people understand our questions and our answers well enough to talk about the same things.

Relationships are important when making complex decisionsWe need teams with the expertise to meaningfully challenge current decisions. We also need teams that can imagine better options and advocate for the policy changes needed to implement them. Like all teams, disaster research teams need mutual understanding and trust, which can only be built through sustained interaction.

Teams need a common platform to coordinate their workThis platform must consider the input of all members and do so transparently enough to get them to support their products. In skilled hands, decision science’s subjective formulation of “expected utility” can provide this platform, capturing the options, beliefs, and preferences of all parties.

By adopting these principles, the Insurance Project was able to identify key economic and psychological differences between flood and earthquake insurance. These differences had practical implications for the marketing and regulation of policies. They also had theoretical implications, which Howard and his colleagues pursued by both conducting psychological experiments based on economic reasoning and conducting economic analyses based on psychological reasoning. The current insurability crisis shows how sadly prescient this research was and how relevant its methods still are.

Over time, Howard and his colleagues expanded their hybrid research approach to other areas, including vaccines, climate change, energy conservation, and workplace safety. In each case, the work was good for society—by applying what we knew—and good for science—by showing what we still needed to learn. In each case, success depended on the intellectual and interpersonal skills needed to build and maintain effective teams. No one had those skills like Howard.

If we’re lucky, the relationships of successful work teams spill over into our personal lives, so that future work also provides an opportunity to meet. Howard was always there in that regard, too. We met at Gil White’s conferences in Boulder, at meetings of the Society for Risk Analysis, in the working groups Howard created, at the workshops he convened at Wharton, and on the trips that brought us closer together.

Howard and his family also supported my family. During a year we spent in Cambridge, England, Howard found an excuse for us to visit IASA, where he was working to prevent further drift between East and West. My children, Maya and Ilya, still remember paddle boat rides on the lake. Howard’s first wife, Sylvia, visited us in Cambridge. To this day, we keep leftover crayons and stickers in a wooden box where she brought us a Sachertorte. We also have fond memories of meals with his wife of 33 years, Gail, including some where we were lucky enough to visit with Laura, Joel and their families.

And Howard is still here, for those lucky enough to have internalized the echo of his voice and energy, posing challenges and inspiring people to face them with him. I can hear and feel that voice and energy as I reflect on the evolution of my own work and the lives we have shared together.