How Do We Measure Risk-Taking in the Brain?

Short answer: There are countless ways we could measure risk-taking in the brain. But one paradigm I find particularly elegant is known as the Balloon Analogue Risk Task (BART).

The Balloon Analogue Risk Task

The BART is simple. A balloon is shown on the screen, and players can inflate it to earn money. (See an excellent video demonstration of the task here). There are two buttons: “Inflate” or “Cash Out.” As the balloon inflates, the reward for cashing out goes up, but so do the chances of the balloon popping. If the balloon pops, no money is earned for that round.

Experimenters typically have people play many rounds of the game to get an average sense of their risk-taking preferences. More inflations on average tells us that someone has either lower sensitivity to risk, higher sensitivity to reward, or both.

Functional Magnetic Resonance Imaging

The BART was originally devised as a psychological experiment to measure risk and reward preferences, but it has since caught on in neuroscience. By having participants play the BART while their brain is being scanned in an MRI (magnetic resonance imaging) scanner, scientists can measure real-time brain activity in response to what happens on the screen.

Functional MRI (fMRI) works by measuring blood flow in the brain using extraordinarily powerful magnets. The magnets in modern MRI machines are strong enough to lift a car. (But they are completely safe, as long as you don’t have any metal on you.)

The oxygen in your blood is ever so slightly magnetic, so when blood flow increases or decreases in certain areas of your brain during a task, fMRI can measure those changes.

However, it’s not as simple as more blood flow means more brain power. Sometimes reduced blood flow can mean increased efficiency, and increased blood flow can mean more struggle during an unfamiliar task.

For these reasons, fMRI studies can sometimes produce conflicting results. Fortunately, a group of researchers from China published a meta-analysis of 22 fMRI studies measuring risk-taking in the brain using the BART. Meta-analyses (studies of the results of many other studies) are the gold standard for understanding many things in science.

Risk-Taking in the Brain

The meta-analysis found remarkable consistency in the neural signatures of risk-taking. Across all 22 studies, significant activation patterns were found in three primary brain networks: the reward, salience, and executive control networks.

In the reward network, researchers found that the right putamen and left caudate areas of the brain all activated more when participants inflated the balloon. These areas were involved in calculating the reward value when deciding to inflate. In the salience network, bilateral insula and dorsal anterior cingulate cortex activity were associated with balloon pumping, likely working in tandem with the reward network in appraising value and focusing attention.

Lastly, in the executive control network, right dorsolateral prefrontal cortex (dlPFC) activity was positively associated with balloon pumping decisions. The authors note that another study found that inhibiting the dlPFC using targeted magnets produced increased risk-taking behavior, indicating that during risk-taking, the dlPFC may be playing an inhibitory control role, making sure the risk-taking does not go too far.

Age-Related Changes

All of these results were consistent across adolescents and adults. However, researchers also noted some age-dependent results. Compared to adolescents, adults showed more brain activity during risk-taking in the right thalamus and midbrain. Conversely, adolescents showed more activity in the reward (bilateral putamen), salience (bilateral insula), and executive control (right dlPFC, left frontal lobe) networks.

Adolescents typically engage in more risky behaviors than adults, including on the BART. Traditional developmental theories argue that this is due to an underdeveloped executive control network (relying heavily on the prefrontal cortex, the part of your brain involved in decision-making that is still developing into the mid-to-late 20s). However, researchers note that the increased dlPFC activity relative to adults could mean that adolescents are in fact engaging in more cognitive control for the same levels of risk, and engage in risk-taking not out of carelessness but deliberate exploration.

Conclusion

Overall, these results are a useful sanity check. You would expect brain areas involved in reward processing, attention, and inhibitory control to be implicated in risk-taking, and that is exactly what researchers find: not just across a single study, but across all 22 studies over the last two decades that have deployed the fMRI and BART paradigms since they were developed.