Bull, bear or one of us?

Photographer: Joel Sartore/National Geographic/Getty Images

This Is Your Brain on Risk

Noah Smith is a Bloomberg View columnist. He was an assistant professor of finance at Stony Brook University, and he blogs at Noahpinion.
Read More.
a | A

How do human beings behave in response to risk? That is one of the most fundamental unanswered questions of our time. A general theory of decision-making amid uncertainty would be the kind of scientific advance that comes only a few times a century. Risk is central to financial and insurance markets. It affects the consumption, saving and business investment that moves the global economy. Understanding human behavior in the face of risk would let us reduce accidents, retire more comfortably, get cheaper health insurance and maybe even avoid recessions.

A number of our smartest scientists have tried to develop a general theory of risk behavior. John von Neumann, the pioneering mathematician and physicist, took a crack at it back in 1944, when he developed the theory of expected utility along with Oskar Morgenstern. According to this simple theory, people value a possible outcome by multiplying the probability that something happens by the amount they would like it to happen. This beautiful idea underlies much of modern economic theory, but unfortunately it doesn’t work well in most situations.

Alternative theories have been developed for specific applications. The psychologist Daniel Kahneman won a Nobel Prize for the creation of prospect theory, which says -- among other things -- that people measure outcomes relative to a reference point. That theory does a great job of explaining the behavior of subjects in certain lab experiments, and can help account for the actions of certain inexperienced consumers. But it is very difficult to apply generally, because the reference points are hard to predict in advance and may shift in unpredictable ways.

Another alternative, developed by various economists, modifies expected utility in certain ways that make it a bit better -- though still not great -- at explaining broad financial market movements. But because the theory relies on people’s expectations of things far in the future, it is very difficult to test in a laboratory or field experiment.

Not only do we not know how people make choices about things like investments, but we don’t even know how they form their beliefs about the future. People don’t seem to follow the kind of mathematically correct reasoning that most psychologists and economists think of as rational. Instead, they use a host of shortcuts and heuristics, and fall prey to various biases. It’s very hard for psychologists to predict which of these will be important in any given situation.

So economists and psychologists have so far failed to find a general theory describing how people respond to an uncertain world. Maybe it’s time for neuroscientists to try their hand at it.

What can neuroscience do to help us understand risk? However humans actually make decisions, it must happen via some mechanisms in the brain. If we can identify those mechanisms, we can watch them in action while people make decisions. That could shed light on all kinds of questions.

For example, neuroscientist Kelly Zalocusky of Stanford University has found a group of neurons that light up whenever a rat is about to choose a safe option over a risky one. If we can find a similar structure in the human brain, then by watching those neurons, we can know -- more or less -- when a person is trying to avoid risk.

That could help answer a number of questions about decision-making in real-world situations. For example, suppose we were to observe the brains of investors right after they lose money. If the risk-aversion part of their brain lights up when they buy stocks after a loss, we know that they see not buying as a risk. That would imply that they expect the market to rebound. But if the risk-aversion neurons don’t light up, that would imply that they became less risk averse after a loss -- just as Kahneman’s prospect theory would predict. That phenomenon is called loss aversion, and predicts that when investors have taken a loss, they become risk-loving and start gambling in order to try to break even. If confirmed, loss aversion could help us understand investor behavior.

A few neuro-economists are already doing these kinds of studies. Colin Camerer, Peter Bossaerts and a team of co-authors used magnetic-resonance imaging to measure people’s behavior during financial-market experiments that approximated the conditions of an asset bubble. They found that when asset prices diverged from fundamental values, subjects showed a brain-activation pattern that might indicate that they were paying more attention to the actions of other subjects. That implies that investors often rely on social cues when making decisions in unfamiliar markets -- a phenomenon that would explain why housing bubbles spread from block to block like a virus.

As brain-measurement technology improves, it will get a lot easier to do experiments like these. That will lead us toward ever-greater understanding of how people deal with risk. Banks, asset managers and the government have a big stake in advancing the state of knowledge, so they should probably think about funding this sort of research. In the meantime, economists and psychologists can look forward to learning which of their theories apply in various situations -- and whether they need to work on some new ones.

This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.

To contact the author of this story:
Noah Smith at nsmith150@bloomberg.net

To contact the editor responsible for this story:
James Greiff at jgreiff@bloomberg.net