Hsu Puts the Brain's Decision-Making Processes Under the fMRI Spotlight
Ming Hsu likes to present people with tough choices. For example:
Which of two coins - one that is perfectly balanced or one that has an unknown bias in favor of either heads or tails - would you pick to best enhance your chances in a coin flip? Your answer on this one most likely reflects how you also make certain investment decisions.
After looking at photos and reading biographies of children living at a real-life children's home in Africa you are asked to decide how to allocate food to them. Don't expect to feel entirely good about your charity, however. Hsu set up the experiment so test subjects have to choose between giving a set number of meals to one child or giving fewer meals, but to more children.
The coin question relates to what is called the Ellsberg Paradox, which has to do with aversion to ambiguity in decision-making, while the children's home dilemma came from an experiment Hsu ran while at Caltech using functional Magnetic Resonance Imaging (fMRI) to study how the brain functions during moral decision-making. As it turns out the odds in the coin flip are the same no matter which one you choose and the children's home gained a lump sum donation for food at study's end. What Hsu gained was insight into how the human brain works when asked to make difficult choices involving primal issues like risk/reward and morality.
Hsu's area of study is neuroeconomics, which is a combination of neuroscience and economics and a subset of behavioral economics. Hsu, who has a Ph.D. in Social Sciences from Caltech, is currently a Beckman Fellow and he will be joining the faculty of the Department of Economics at Illinois in the fall as an assistant professor.
"It's combining neuroscience and economics," Hsu said of his field. "Traditionally economics is formal models of decision-making, and some of it is from real-world data. Behavioral economics is sort of a fusion of psychology and economics, and neuroeconomics brings in neuroscience as well."
Neuroeconomics is a cutting edge field in a discipline, economics, with a history of being slow to adapt to new ideas (using fMRI techniques to study economic decisions is less than a decade old). Nevertheless, and not surprisingly, the idea of looking at the brain as it deliberates questions like how to distribute food to needy children or how to invest one's money has drawn a national spotlight. Hsu's work at Caltech has been written about in the Wall Street Journal and was featured in a recent episode of the PBS science show Curious, aired nationally on Jan. 24.
That show focused on the experiment run by Hsu and his colleagues at Caltech involving the Canaan's Children's Home in Uganda. Test subjects were put into a headscanning magnet and made donation decisions by pushing one of four colored buttons to move a ball on a screen to one of the donation options. Hsu calls it a distributive justice experiment.
"They have a decision whether to donate more money in total to the kids or to be more equitable in their distribution," Hsu said. "So you can imagine 'I can give 24 meals to one kid or 10 each to two.' In one case you are giving more in the sum and then the second case you are being more equitable and you are giving to more kids, but the total is less."
Hsu said they varied the donations so that there was a tension between being equitable and donating the largest amount possible to the children.
"We can vary these numbers to see what your aversion to inequity is and see what brain areas are responding to issues of equity and efficiency," he said.
That ability to look at the brain while it processes decisions about issues such as equity and risk is providing new insights not only for economics, but also for neuroscience, psychology, and even philosophy.
"This is about a very specific subset of moral decisions," Hsu said. "This goes back to a lot of the political philosophers who talk about how we allocate burdens and benefits in a society. In formal terms it's called distributive justice. There are a lot of theories about how we should do this, and our study comes from an economic approach. We study that within the context of individual decision-making."
Neuroeconomics as a branch of behavioral economics investigates biological and neural processes as people make decisions about purchasing, investing, or bargaining. And Ming is a young researcher at the forefront of this developing field. While at Caltech, he was lead author on a paper about the Ellsberg Paradox, Neural Systems Responding to Degrees of Uncertainty in Human Decision-Making that was published in Science in 2005 and he has done a paper about the experiment with the children's home for Science in 2008. Although he uses technology like fMRI and views topics through the lens of neuro-science, Hsu said his field of neuroeconomics fits within traditional economics.
"Right now what we've done is coming from a very traditional economic perspective," Hsu said. "I think only a really hardcore rational choice economist would be surprised by that. I think it's just filling in the picture. The interesting things are things we never thought about before like the connection between inequity and risk."
Those connections are at the heart of the neuroscience aspect of Hsu's research. The headscanner measured blood flow during both the children's home experiment and the study about the Ellsberg Paradox, illuminating which brain regions are activated when people are asked to make those kinds of decisions.
"We found basically that the classical decision regions are very much involved in this type of decision making," Hsu said. "In the past people haven't really found that. We found in the striatal regions in the dorsal striatum that encode equity and efficiency, the more money you give to these kids and the more equitable it is, the more these regions fire. These regions are the ones that are activated when you get money personally. Rats are happy when cocaine is injected into these regions.
"One of the emotional regions in the brain, the insula, responds to inequities, so the more inequity that the subjects are able to save with their choices, the more this region fires. It's almost as if this region is pushing you toward making the more equitable choice. The difference between subjects' activation of the insula have in other studies been found to correlate with their level of risk-aversion."
Hsu said issues of equity and risk are connected in the brain.
"You can relate equity sensitivity or moral sensitivity to brain function," he said. "Both inequity and risk have very deep theoretical connections. With those you are essentially comparing distribution. If you think of a financial market, it's a distribution of returns that you are worrying about when it comes to risk. With equity you are caring about the distribution of, say, income or education or whatever variable you want to measure in a population."
Any investigation that could shine a light as to why people make economic decisions would, needless to say, be of interest to those concerned with finances and investments. In addition to the Wall Street Journal the work of Hsu and his colleagues has been written about in Forbes and the Chicago Tribune. His initial project using fMRI, which resulted in the paper on the Ellsberg Paradox, had ramifications for financial planning.
"The basic idea is that there are different types of uncertainty in the world," Hsu said. "In classical theory of uncertainty you have some probability distribution over some event, and you use that information. The classic counter argument is the Ellsberg Paradox."
Hsu said most people in the coin flip experiment usually choose the unbiased coin, even though the odds are the same as with a biased coin (again assuming you don't know the bias) as a balanced coin. The Ellsberg Paradox experiment tested what brain regions fired when people were evaluating ambiguous gambles versus risky gambles where the probability is known.
"Most people would prefer to bet on the known fair coin, even though you can say for all objective reasons it's the same as if you chose the biased, unknown coin," Hsu said. "There has been a lot of work in the last 30 years, generalizing this notion to explain this behavioral aversion to what's called ambiguity to real-world phenomenon."
As an example, Hsu pointed to financial investments. Since stocks are seen as riskier than other investments such as government bonds, investors expect high returns - more risk, more reward - but they are more likely to take that risk than, for example, the ambiguity of investing in an unfamiliar country.
"If you are investing in a stock market, would you rather invest in your home country, in U.S. stocks, or would you rather diversify all over the world," Hsu said. "If you diversify all over the world it's a good thing because every banker says you should diversify your portfolio. But of course people don't do that, people invest overwhelmingly in their own country. One of the explanations is that people have this ambiguity aversion and invest in their own country because they feel like they know more about it and they feel like it's just safer, even when it's actually not a good thing to do."
The topics of risk/reward and equity that Hsu explores are intimately related to our brains' evolutionary heritage. Risk/reward factors that have developed in humans and other animals over thousands of years while foraging for food or keeping an eye out for predators are deeply ingrained and play roles in decision-making for all sorts of things. Hsu said studying those factors through fMRI simply adds to our knowledge of how and why people make choices.
"People respond to incentives; I don't think you'll find any kind of behavioral economist or neuroeconomist who will disagree with that," Hsu said. "We know that institutions matter, we know that social behavior matters. It's really about the way that the brain implements (behaviors) and how social institutions change those behaviors. So a lot is filling in a more complete picture of how people actually make decisions."
Hsu believes there is plenty of room to explore these issues at his new home, including in an area that several Beckman psychology researchers focus on: the aging mind.
"One of the nice things about the University of Illinois is that there is a very strong aging group here," he said. "When people are saving for retirement those (topics) are things that we know people are extremely biased for in their actual decision- making. I think part of the reason is that evolution has designed us to just not think about what is going to happen when we're 80 years old.
"So there is a lot of work in what kind of decision processes lead to under-saving by older people and how we can change the framing of the problem to get them to save more and what kind of therapies you can use to get very old people to improve their decision-making."
Despite being a researcher who has yet to start his faculty career, Hsu is already crossing disciplines from psychology to economics to philosophy and earning national recognition.
"These are questions that every notable philosopher has asked since basically the beginning of philosophy," he said. "To the extent that you can bring some new ideas into it, people are interested."
"Christine", one of Ming Hsu's subjects in a moral decisionmaking experiment. Image courtesy WNET Thirteen.
The choices required of test subjects in Ming Hsu's moral decision-making experiment were not easy, according to several of the study's subjects. The PBS show Curious, produced by WNET Thirteen in New York, followed Hsu and his fellow researchers as they ran test subjects through a headscanner and asked them to make decisions about donating to a children's home.
Watch an online version of the Curious episode, titled Mind, Brain, Machine.
This article is part of the Spring 2008 Synergy Issue, a publication of the Communications Office of the Beckman Institute.