In This Issue
Winter Issue of The Bridge on Complex Unifiable Systems
December 15, 2020 Volume 50 Issue 4
The articles in this issue are a first step toward exploring the notion of unifiability, not merely as an engineering ethos but also as a broader cultural responsibility.

Embracing Behavioral Science

Friday, December 18, 2020

Author: Leidy Klotz and John Pickering

A defining feature of complex systems is that fully predicting the effects of changing them is impossible. Thankfully, engineers have never been deterred by the specter of impossibility.

When it comes to designing in complex systems, however, there remains an obstacle to unleashing the full force of engineering. The problem is that an essential science for understanding and influencing complex systems is deemed by engineering to be either out of scope, or—at best—an optional consideration for niche disciplines. This needs to change. Because, whether the goal is to quickly introduce a new vaccine or to stabilize climate change, a unifying characteristic of complex systems is that they are driven by human behavior.

Role of Behavioral Science in Engineering

Asking engineering to embrace behavioral science is not unreasonable. In fact, behavioral science provides the overlooked foundation of modern engineering. Sure, engineering often involves creatively applying science and math. But how this science and math are creatively applied rests on assumptions about logic and reason.

It is assumed, for example, that engineering will derive and then select optimal solutions given a set of constraints and goals. It is assumed that engineering will be consistent in how these goals are set and in how resources are valued. Such basic assumptions about engineering design do not garner much attention. They are, after all, “common sense.”

But behavioral science has shown that some of these foundational assumptions are wrong. People satisfice, settling for good enough solutions and leaving optimal ones unexplored. People value the exact same thing differently depending on whether they have this thing or not. Such systematic deviations from old models of logic and reason better explain the behavior of complex systems—and can better inform design within these systems.

Mental Traps in Engineering

To take full advantage of opportunities for behavioral design in complex systems, engineering needs to overcome a couple of mental traps that have held back its fields, and therefore society.

To overcome the first mental trap, engineering should stop thinking of behavioral science as “soft,” which too often connotes a lack of rigor. Certainly, human behavior varies with context. The degree to which people satisfice, for example, is not absolute. It depends on how much people care about the problem at hand.

But the behavior of molecules and materials also varies with context. Newton’s apple falls at a different rate through water, or with a parachute attached. The need to adjust scientific generalizability to practical context is why there are so many variables and coefficients in engineering equations. Precious little behavioral science has been formalized into variables and coefficients, and that is all the more reason for engineering to engage.

To overcome the second trap, engineering should stop concerning itself with the hierarchy and prestige of science. Complex systems trample past the made-up boundaries between disciplines, fields, and types of science.

Contrast the reality of complex systems with how engineering programs are accredited, with students required to learn “sciences appropriate to the discipline” and sciences narrowly defined as “biological, chemical, and physical.” Simply eliminating this latter distinction and allowing that engineers can benefit from all sciences would go a long way toward producing engineers who more fully understand the logic and reason that underpin their disciplines—not to mention complex systems.

Engineering researchers must overcome these mental traps in order to engage in research that needs the perspective of engineers. For example, to rationally consider the prospects of climate engineering, we need to understand how engineering behavior aligns with and deviates from findings observed across human populations.

If we are serious about engineering in complex systems of any sort, we need to acknowledge how these multifaceted and unpredictable systems change the cognitive load on designers. In other words, what mental shortcuts do we bring to engineering complex systems? Which are harmful? Which are helpful? And we need to consider engineering-specific social norms to find ways to dismantle systemic sexism and racism in the field.

Engineering practitioners who overcome these mental traps will be rewarded with an expanded role in creatively applying science. Because behavioral science is already being applied, with or without the aid of engineering. Listen to the Nobel-winning behavioral scientist Daniel Kahneman: “There is a technology emerging from behavioral [science]. It’s not only an abstract thing. You can do things with it.”[1]

Listen to the market: hundreds of organizations and companies have scrambled to build capabilities in behavioral design. The question is not whether behavioral science will be applied. The question is whether engineering will play a leading role.

Closing Thoughts

Engineering will sacrifice impact and market share, and complex systems will remain more vexing than necessary, until engineering embraces behavioral science. There is no time to wait. Minimizing pandemic deaths depends on the behavior of those least at risk; removing carbon from the atmosphere requires different attitudes from those that put it there; and maintaining social growth within planetary boundaries requires simultaneous consideration of environmental impact and human wellbeing.

Our common future depends on our ability to understand and influence complex systems, which depends on our success in creatively applying behavioral science.


[1]  Edge Master Class 2008: Richard Thaler, Sendhil Mullainathan, Daniel Kahneman – A Short Course in Behavioral Economics, Jul 25.

About the Author:Leidy Klotz is the Copenhaver Associate Professor at the University of Virginia. John Pickering is cofounder and CEO of Evidn.