Verified by 
Even if you don’t go to casinos, you regularly have to make bets: on investments, on job decisions, on whether you will need an umbrella on Tuesday. Three main factors determine how we make such decisions. What are the risks? What are the rewards? And what is your internal bias—the accumulated experience that nudges you in one direction or another? Been on a losing streak? Made a lot of money recently? Got caught in the rain last week? All of that matters, which is why economist Richard Thaler won the Nobel Prize for developing “nudge theory.” Now in an unusual new study, a team of neuroengineers from Johns Hopkins University, shows for the first time how that bias plays out in our brains, millisecond by millisecond.
It turns out that, like an angel and a devil sitting on opposite shoulders, the two sides of the brain engage in a tug of war. The right hemisphere pushes us toward risk and the left pulls us away from it. And while making risky decisions, we don’t just use the parts of our brains that handle reason and judgment. Our decision-making reaches deep into the brain to areas associated with emotion such as the amygdala.
This research advances our understanding of how decision-making is encoded in the brain and might refine therapeutic treatments for gambling addiction or for people with psychiatric and mental disorders such as Parkinson’s disease. Treatments like deep brain stimulation (DBS) already work by changing the brain patterns in Parkinson’s patients. This new kind of manipulation, should it be developed, would add treatment for impaired decision-making.
More controversially, says Sridevi Sarma, the senior author on the new paper, which was published in Proceedings of the National Academy of Sciences, “you can potentially control a person’s decisions by making them take more or less risk.” As an example, she points to military commanders who might want to increase the soldiers’ willingness to head into danger.
Obviously, the ethics of that would have to be debated. And the study led by Sarma and Pierre Sacré, who was a post-doctoral fellow in Sarma’s lab and is now at the University of Liège in Belgium, is not one that can be easily recreated. Its participants were ten epilepsy patients who had electrodes implanted in their brains in order to locate the origins of their seizures. (In severe cases, neurosurgeons operate to remove that brain tissue to stop seizures.) That set-up allowed Sarma and Sacre and their colleagues to track neural activity throughout the brain in real time, meaning milliseconds. Other techniques don’t allow such precision or such wide coverage of the brain.
First, the researchers had to work out how to estimate the bias that each individual brought to each decision. To do that they set up a gambling game on a computer. It included an unlimited deck of only five cards: 2, 4, 6, 8 and 10. One card was face-up (the participant’s), the other face-down (the computer’s). Participants had to bet ($5 or $20) on whether their card was higher. That’s easy with 2s and 4s (you’re likely to lose) and with 8s and 10s (you’re likely to win), but on 6s, when you are equally likely to win or lose, people do “a variety of weird stuff,” Sarma says. In addition to weighing risk and reward, each person’s internal bias comes into play. “How you feel when you gamble in a casino is based on past outcomes.” With a prediction of bias in hand, the researchers were able to compare it to the readings from the electrodes in the brain to ask, “what part of the brain is modulating and moving with or against this internal bias?”
They found that the brain uses what they call a “striking” push-pull phenomenon. “Right hemisphere is pushing you to take the bet, take the risk, and left hemisphere is pulling you away from that,” Sarma says. According to Sacre, “there’s no clear answer for why we see these lateralizations in different brain functions,” but they exist in other types of brain processing such as the instinct to approach or avoid. “This push-pull phenomenon seems to be evolutionary,” Sarma says.
It was already known that, on average, people take more or less risks if the left or right sides of their brains were stimulated. But until this study no one had tracked the way that bias shifted with each subsequent bet, i.e., trial by trial in the experiment.
There are plenty of limitations here. As noted, the study included only ten patients and they were all epileptic. Some critics worry that skewed the data, though Sarma and Sacre believe these patients are otherwise healthy and they controlled for some of potential problems. They knew, for instance, which parts of each person’s brain triggered seizures and did not include those regions in their analysis. But they also argue there is no other way at this point to capture these types of recordings since you cannot put electrodes in a human being’s brain unless it is clinically warranted.
In the future, Sarma and Sacre plan to explore whether non-invasive brain stimulation, such as transcranial magnetic stimulation (TMS), would have the same effect. And now that they have identified the push-pull effect, they could explore it simply by rigging the card decks in their experiment—no stimulation required.
“This is just the tip of the icerberg,” says Sarma. “As soon as you understand how the brain governs behavior, then you can manipulate it.”
