Using Behavioral Science to Redesign the Built Environment

China’s most famous painting is “Along the River during the Qingming Festival.” Often referred to as “China’s Mona Lisa” (more for its fame and mysterious history than for any likeness to da Vinci’s portrait), the painting dates from the early 1100s and stretches over 17 feet. In intricate detail, the painting depicts daily life in Bianjing, now the city of Kaifeng. You can see farmers, merchants, soldiers, and commoners going about their lives inside the city and along the river and mountains.

I was first introduced to “Qingming Festival” by Wente Pan, a visiting Ph.D. student in our behavioral science for sustainable systems group at the University of Virginia. In our first meeting, Wente presented me with a full-size replica of the painting. As I unfurled the scroll, I was first amazed by the length and then moved by the vivid detail, it’s hard not to be. But something else about “Qingming Festival” really caught my attention.

As a professor of civil engineering and architecture, I was impressed by how modern Bianjing seemed. The buildings appear welcoming and durable, with roofs to keep water out and windows to let light in. They’re also organized thoughtfully, with public gathering places between them and roads connecting the homes, offices, and stores to one another.

But the fact that the city looks so similar to the present day also left me wondering: Shouldn’t engineers, planners, and architects like me have contributed to a vastly superior built environment from the one that existed nine centuries ago?

Shouldn’t engineers, planners, and architects like me have contributed to a vastly superior built environment from the one that existed nine centuries ago?

Certainly, I’m glad I live in modern Charlottesville, Virginia, and not in Bianjing in the 1100s. We have two bathrooms, and they are inside. We have heating that doesn’t require constant fueling, nor does it spew smoke into our home. We have smoke detectors and fire-resistant materials. Yet, these improvements are modest compared to the innovations in other domains, like aviation and computer science, over the same time. Imagine if one of the carriage drivers in “Qingming Festival” was transported to a modern city. I suspect the time traveler would recognize the buildings by their function but would have no idea what to make of airplanes or smart phones.

I believe similar innovations in the built environment can arise from integrating behavioral science, engineering, and design. Moreover, innovations in the built environment are critical to ensuring a sustainable future.

Buildings and transportation alone, for example, directly account for nearly half the energy use and carbon dioxide emissions worldwide. Over 800 billion tons of precious natural resources are currently “used” in the built environment. This extraction of natural resources, and the associated waste and emissions, are making our environment less hospitable to human life, often in irreversible ways. With this is mind, it’s not hard to see why two of the United Nations’ 17 Sustainable Development Goals (“Sustainable Cities and Communities” and “Innovations in Infrastructure”) focus on the built environment.

The intersection of behavioral science and user interaction with the built environment has produced engineering advances that address sustainability challenges. Consider smart thermostats that automatically adjust to occupant behavior patterns or heat sensing lights that turn off when no one is in a room. Such systems can save energy and improve the occupant experience and are not limited to thermostats. In their review of research on user behavior and residential energy use, Charlie Wilson and Hadi Dowlatabadi describe a range of ways behavioral science help explain how people interact with the built environment—including choices for one-time weatherization projects and repetitive energy consumption activities such as lighting and appliance use.

However, we must extend behavioral insights for the built environment from users to designers. Transformative innovation happens long before the first room is occupied. Sure, default settings on thermostats can prevent users from wasting energy cooling their homes when no one is there. But what about using defaults that encourage designers to consider passive cooling techniques that eliminate the need for mechanical air conditioning altogether?

Such work applying behavioral science to built environment design is in its infancy, but it does show the potential benefits of a greater focus on this area. For instance, Thomas Beamish and Nicole Biggert’s sociology-inspired research on construction organizations finds widespread evidence of what they call a “default design heuristic,” a tendency to overuse designs from prior projects. For example, they discover an industry default design for a commercial office building that may sound familiar: two or three stories; 50,000­–65,000 gross square feet; rectangular; windowed premium offices on outside; interior spaces for office cubicles; and parking adjacent to the structure. Such a heuristic could partially explain why the buildings in 12th century Bianjing don’t look much different than buildings today.

We must extend behavioral insights for the built environment from users to designers. Transformative innovation happens long before the first room is occupied.

Behavioral science promises not only better understanding of built environment design but also ways to improve it. My colleagues and I have shown how various types of choice architecture interventions, such as framing effects, endowed defaults, and role model examples, encourage more ambitious sustainability goals among students and professional designers. For example, designers using the Envision® rating system for sustainable infrastructure set around 20 percent more ambitious goals for sustainability when they are given points to lose for poor performance as opposed to the current system, in which they begin with no points and gain them for good performance. Scaled to all U.S. infrastructure, 20 percent better performance eliminates over 3 billion tons of CO2 (estimate based on a per-capita carbon footprint of infrastructure of 53 tons and a U.S. population of 316 million).

Of course, infrastructure is not updated all at once. But considering the successful cash-for-clunkers program invested around $3 billion and saved an upward estimate of 30 million tons of CO2, tweaking defaults in an infrastructure rating system appears relatively promising.

To translate these limited research examples to the widespread innovation in our built environment, we must overcome artificial barriers between behavioral science and engineering. Engineering is supposed to be the creative application of science. And yet engineering still relies almost exclusively on older physical (e.g., physics, chemistry) and formal (e.g., mathematics, logic) branches of science. Other applied disciplines (e.g., public policy, medicine, education) are ahead of engineering in integrating behavioral insights. Similar progress in engineering would add a new dimension to the discipline.

Opportunities for research abound. I promise to work on the engineers, demonstrating ways to extend our approach to creatively applying science can incorporate behavioral science. Behavioral scientists can help by talking to engineers (at least the extroverted ones who look at your shoes when conversing) who will be excited to share the nuances of the high-impact, long-lasting decisions most in need of behavioral insights.

Together, we can find ways not only to convince users to turn down their thermostats in the winter but also to enable designers to create comfort that renders fossil fuel-powered heating unnecessary. Around 900 years from now, when people look at pictures of present-day Charlottesville, we can ensure they find it more of a relic than I found Biajing.

Further Reading & Resources

  • Shealy, T., Klotz, L., Weber, E. U., Johnson, E. J., & Bell, R. G. (2016). Using Framing Effects to Inform More Sustainable Infrastructure Design Decisions. Journal of Construction Engineering and Management142(9), 04016037. (Link)
  • Beamish, T. D., & Biggart, N. W. (2012). The role of social heuristics in project-centred production networks: Insights from the commercial construction industry. Engineering project organization journal2(1-2), 57-70. (Link)
  • Wilson, C., & Dowlatabadi, H. (2007). Models of decision making and residential energy use. Annual review of environment and resources32, 169-203. (Link)