The economics and psychology departments at the University of Illinois teamed up to learn how people think competitively. Their findings help explain strategic behavior and potential causes of neurological disorders; their experience provides a taste of what it is like to conduct multidisciplinary research.

Belief learning is the process of learning the intentions of others, usually during some form of competition. In other words, it is what you believe your opponent is going to do.

“Lots of people study classical learning, about how I just turned right and got food, so I’m going to turn right again,” said Kyle Mathewson, postdoctoral fellow at the University of Illinois, “but this is a different beast. This is trying to learn about the intentions of other people.”

So instead of behaving in a way that will result in the greatest reward, it is about anticipating the actions of opponents so you’re more likely to win. Unlike classical conditioning, which was discovered by enticing dogs to drool, this kind of learning is seen mostly in creatures capable of higher thought, since it involves speculative reasoning where competitors try to predict the actions of others and calculate their own course of action.

Belief learning can be applied to the competitive behavior of firms and to many everyday situations. In the study, subjects participated in a “patent race,” which was meant to represent the competition between firms and businesses.

During this game, players were encouraged to invest a certain amount of money to beat the other person, where the person who invested the greatest amount of money won the patent. Ideally, players are aiming to invest the smallest amount possible to win; players went all-in to guarantee success over rivals.

It may seem like an easy game. But it’s difficult to win because the players are learning what to expect from their opponents. Losing players expect the person in the lead to go all-in, so they stop investing and save their money. After awhile, the person in the lead will realize nobody is investing and begin to spend less. This behavior still keeps the game alive because players need to anticipate when such changes in their opponent’s behavior will occur. This type of learning is seen in any kind of competition, in which players need to anticipate their opponent’s strategies.

Beside providing a better understanding of human behavior, this study also provides information that can be used to improve mental health. While the subjects played the patent race game, their brain activity was studied.

“People damage their brain and are normal but lose the ability to tell when people are trying to trick them,” Mathewson said. “Once we understand how the brain computes this information, how it keeps track of these calculations and how it helps you to learn to interact with people, then we can understand how it goes wrong.”

For the researchers, the main learning experience was working with other departments. The biggest obstacle was learning how to communicate because two fields can have different terminology for similar phenomena. For example, the concept of belief learning in psychology is a part of game theory in economics. To help overcome the language barrier, they thought of research ideas based on better understood topics.

“We chose learning, because it has been studied for a long time. So it has concrete, well-defined definitions, whereas the terminology in other areas of psychology are still being formulated,” said Ming Hsu, professor at the Haas School of Business and Helen Wills Neuroscience Institute at the University of California, Berkley, whose team collaborates with the researchers at Illinois. The differences in jargon not only made the collaboration between psychologists and economists confusing, but also limited what topics the interdisciplinary research team could study, because they needed to stick to more defined areas of psychology.

Additionally, the research team had to face other obstacles unrelated to the theory behind their research. One such issue is finding subjects, because people usually do not respond kindly when asked if their brains can be watched. To overcome this, Hsu said the researchers needed to provide some incentive, usually in the form of money. Often the subjects don’t care about the topic of the research once they are offered money to participate, he added.

The research team discovered that the ventral striatum and rostral subcallosal ACC, the parts of the brain primarily engaged in the activation of behavior, have enhanced activity during competition. This discovery will lead to further research identifying the jobs of certain regions of the brain.