Decision Neuroscience

By Tessa Shaw

Traversing disciplinary boundaries to advance our understanding of financial decision making.

Sitting at his desk in the City of London about fifteen years ago, Carsten Murawski never imagined that one day he would be looking at images of brains on his computer screen. As an analyst at a leading US investment bank, his world revolved around mergers, acquisitions and initial public offerings of the world's largest companies. His job was to help hyper-rational economic agents chase opportunities for returns, and his computer screen was filled with spreadsheets of accounting data and stock prices. 

I've always been fascinated by financial markets. But what really fascinates me is the people in those markets. I have always wanted to know how they tick.
-Carsten Murawski

After a few years at the coalface of financial markets, he turned his back on the world of high finance, to explore these markets through research, and later, through science. While working on his PhD at the University of Zürich, he witnessed the beginnings of an exciting new development in finance. Until then, finance research was focused primarily on financial instruments such as derivatives and institutions—mainly, stock markets and companies. Little research existed on individuals, and the models of actors in financial markets were highly abstract, allowing for little variation in individual behaviour. They treated individual decision-makers as 'black boxes', describing what they choose, but not how they choose it.

In the late 1990s, new developments in human brain imaging enabled researchers to measure brain activity in a non-invasive way. Economists at the University of Zürich began to utilise this technology to investigate decision-making, connecting economic theories with the underlying neurobiology, establishing links between variations in economic behaviour and biological variations of the decision maker. Over the past few decades, researchers around the world have meticulously documented that humans systematically deviate in many ways from the workhorse theories of academic finance. Evidence shows there is a large degree of variation in behaviour, not just between individuals, but also within individuals over time. Evidently, a lot of this disparity in economic behaviour is related to biological variation.

Today, researchers like Murawski believe that a better understanding of this variation is necessary, not only to improve the financial behaviour of individuals, but also to improve financial markets. To do so, a better understanding of the processes through which financial decisions are made is required.

Soon after joining the Department of Finance at the University of Melbourne in late 2007, Murawski started using laboratory experiments to study a certain aspect of financial decisions: the trade-off between reward size and delay to reward. For example, would you choose to receive $20 now, or $50 later? This is often referred to as intertemporal choice. His main interest at the time was the role context has in making such decisions—what are the factors that will cause someone to alter their choice from $50 to $20? Standard theories assume that decision-makers only consider information that is immediately relevant to the decision, such as the amount of cash received or the delay to receiving it. But people usually make their decisions in a rich environment in which they are constantly bombarded with seemingly irrelevant information.

This indicates that the brain automatically extracts various kinds of information from visual stimuli in the environment, some of which we never become aware of, but which may indeed be powerful enough to affect unrelated decisions thereafter.
-Stefan Bode

In an early study, Murawski and his colleagues found that even brief exposure to images that people found rewarding (such as the Apple logo)—and that ostensibly were unrelated to the particular financial decisions in their experiment—could systematically influence subsequent financial choices. This signified that an environment brimming with irrelevant information could, in fact, affect the general state of one's reward system, allowing for bias in the encoding of choice options, resulting in choices being shifted. This evidence that financial decisions could be affected by seemingly unrelated environmental factors was a notion that hitherto was outside of standard economic theories.

Having found that they could systematically shift decisions by manipulating the decisionmaker's environment, the researchers then wanted to determine how this happens in the brain. They teamed up with a group of brain imaging specialists from the Florey Institute of Neuroscience and Mental Health in Parkville, one of the world's leading neuroscience research institutes. Among them was Stefan Bode, a young cognitive psychologist. During his graduate studies at the Max Planck Institute for Human Cognitive and Brain Sciences in Leipzig, Germany, Bode used sophisticated statistical techniques, which allow for the prediction of decision outcomes from studying patterns of brain activity as measured using functional magnetic resonance imaging (fMRI). Using these methods, the team was able to identify brain regions in which the representation of choice options changed due to the processing of briefly presented (16 milliseconds) rewarding images. This means that activity in these brain regions allowed them to predict how impulsive the financial choices that followed would be.

This insight further piqued the team's curiosity. They now wanted to discover the type of visual stimuli that could affect financial decisions, and whether people are affected in different ways by their environment. Bode, Murawski and Daniel Bennett, a Psychology PhD student, used electroencephalography (EEG) to measure the brain activity of participants who were exposed to a variety of positive images. While the images were displayed (this time for 3.2 seconds) in the background, the participant's attention was directed to an unrelated task on the screen, comparable to a simple computer game. This scenario is similar to everyday situations in which people passively consume advertisements or other stimuli in the background. At the conclusion of the experiment, participants were shown the images and had to rate them on a number of dimensions, including how arousing they found the individual images to be, or to what extent the images made them think of the present or of the future. It is important to note that the participants were not aware that they would be asked to rate these images at the end of the experiment. Using the same statistical techniques as in their previous study, the researchers found that they could indeed predict people's attitudes towards these images from the brain activity recorded earlier.

"This is interesting because we are constantly exposed to some kind of stimulation in our environment, but how and under what circumstances our decisions are influenced by the different stimuli we encounter, we don't know yet."

These findings have substantial implications beyond financial decision-making, as they could help advance our understanding around unhealthy choices. For example, in a joint project with Annette Horstmann at the Max Planck Institute in Leipzig, Germany, the team found that decision-making in obese people is more susceptible to environmental influences than in lean, healthy individuals. In the hope of using these insights to improve decision-making in highrisk populations for obesity, Bode and Murawski recently partnered up with Melanie Wakefield at the Cancer Council of Victoria to develop new obesity prevention strategies in Australia.

Today, this research at the intersection of economics, psychology and neuroscience is referred to as decision neuroscience. Its aim is to provide a biological foundation for our understanding of decisions, including economic and financial decisions. Such research is highly complex and requires the close interaction of specialists from different academic disciplines.

In 2011, Bode, now in the Melbourne School of Psychological Sciences, and Murawski set up the Decision Neuroscience Laboratory, a cross-discipline partnership between the Department of Finance (Faculty of Business and Economics) and the Melbourne School of Psychological Sciences (Faculty of Medicine, Dentistry and Health Sciences). Bode and Murawski, together with PhD students, work on fundamental aspects of decision-making such as learning about uncertainty, the perception of time, and intertemporal choice. They work closely with other laboratories and institutions, like the Cancer Council of Victoria, allowing their fundamental research to inform solutions to economic and health problems including retirement savings, obesity, pathological gambling and certain mental illnesses such as obsessive-compulsive disorder.

Last year, Peter Bossaerts, one of the pioneers of decision neuroscience, joined the Faculty of Business and Economics. In the 1990s, at the California Institute of Technology (Caltech), Bossaerts was one of the first to use laboratory experiments to study financial decisions and markets, later becoming part of a small group of researchers who started using brain imaging to study human decision-making. His initial work was centred on the encoding of various aspects of risk in the human brain. He conducted numerous ground-breaking studies in this area, many of which inspired new research within finance, neuroscience and psychiatry. Bossaerts, together with colleagues at Caltech and Yale, has also been uncovering the neural basis of theory of mind—our ability to attribute mental states to oneself and others, a skill vital for social interaction. Deficits in theory of mind have been associated with a range of mental illnesses including autism spectrum disorder, attention deficit hyperactivity disorder and schizophrenia.

...this research at the intersection of economics, psychology and neuroscience is referred to as decision neuroscience.

Recently, Bossaerts and his collaborators investigated how a person's development of theory of mind affects behaviour in a financial market. They found that people with more developed theory of mind—that is, people who are better at reading other people's intentions—are more susceptible to riding a bubble in financial markets. This shows that a skill highly beneficial in our day-to-day interaction with other individuals—and indeed a defining aspect of our humanness—can, in fact, work against us in financial markets. It also shows that decision neuroscience not only improves our understanding of individual behaviour but also provides important insights into the behaviour of markets.

"Ultimately, economic outcomes are the result of actions by individuals. And these individuals are biological organisms, with bodies, minds and brains," says Murawski.

Murawski further notes, "For a long time, the research strategy in economics and finance was to abstract from the organism, and even from the individual. While this strategy has been successful in some respects, it has failed in others. There are many important financial phenomena we don't understand, like bubbles and crashes, and gigantic financial problems for which we have effectively no feasible solutions, such as the retirement savings gap. I believe that the key to solving many of these problems lies in a better understanding of how individuals make decisions. And decision neuroscience provides the appropriate tools to gain this understanding."

For further information visit
dlab.unimelb.edu.au