Everyone from seasoned hedge fund managers to Redditors tries to predict which stocks will spike next. According to new research, they’re going about it all wrong.
Previous studies have explored how, when aggregated from a group of people, activity in the brain can predict which videos will go viral, which news articles will be shared, and how people in the real world will respond to advertisements. The practice has been dubbed “neuro-forecasting.”
In a study published Monday in the Journal of Neuroscience, scientists at Stanford University applied this theory to the stock market.
“It turned out in this research that brain activity provided a better forecast than behavior,” co-author Brian Knutson, a professor of psychology and neuroscience at Stanford University, tells Inverse.
How they did it — The researchers focused on two ancient parts of the deep brain: the nucleus accumbens and the anterior insula.
During two experiments, the team took rapid three-dimensional snapshots of 73 participants’ brain activity in these two regions. This snapshot occurred while the participants were deciding whether or not to bet that a stock's price would increase.
Participants made their choices based on a graph they were shown of the stock’s past performance. The team noted how much activity increased in the nucleus accumbens and anterior insula just before participants made their choices.
By combining this activity with the choice each participant made, the scientists found activity in the anterior insula was associated with a person saying “no” to a stock, or avoiding risk. On the other hand, if activity increased in the nucleus accumbens, people were more likely to say “yes” to taking a risk and betting that the stock would increase.
Scientists believe activity in the nucleus accumbens is associated with excitement and taking risks, whereas activity in the anterior insula is associated with anxiety and avoiding risk.
On its own, one person’s brain activity could not predict whether the ticker would go up or down. However, when the team aggregated individual choices, they found that when the majority of people in the group displayed increased activity in the anterior insula, which is associated with avoiding risk, this reliably forecasted when a stock’s price actually went up or down the next day in the real world.
Notably, the team was not able to forecast stock dynamics using conventional stock indicators or participants’ individual investing choices, only with aggregated brain activity.
What’s new — According to Knutson, the finding that brain activity, but not individuals’ behavior, can forecast stock price movements makes sense from an evolutionary standpoint — but it goes against the common scientific idea that one person's behavior should forecast what others will do.
“That’s the weird part.”
This finding turns both behaviorist psychology and some economics theories on their heads.
“We measured their choices and the brain activity in those regions before the choice was made,” Knuston explains.
“That gave us the data we needed to not only forecast their behavior but forecast what the ticker would do next. That’s the weird part. We’re not just trying to predict each individual’s behavior in an experiment, but we’re taking the average of their brain activity and their choices and forecast essentially what a bunch of other people in the stock market did [and therefore whether or not the price would go up or down].”
Why it matters — Because the research is still in its infancy, Knutson says it’s likely not the end game. Instead, scientists may one day find other behavioral or physiological markers that are correlated with the brain activity documented in this study, and try to use those to make forecasts about market trends.
“We have been working on neuroeconomics for 20 years now. I never thought it would get this far,” he says.
This isn’t the first study like this that Knutson has worked on. In a study published in March 2020 in the journal PNAS, Knutson was part of a team of researchers that used activity in the nucleus accumbens and anterior insula to predict how long people would watch a video. They were then able to predict which videos would become viral.
According to Knutson, his team sees the new research as a demonstration that raises a lot more questions for scientists to answer, and one more study that builds on a growing body of research that uses previously hidden brain activity to predict trends.
“We think it can help people make better choices,” Knutson says. The research could also help social scientists better understand how individual choice scales to collective choice.
“This can have a lot of implications, not only for things like internet markets but also for policymakers.”
Abstract: Successful investing is challenging, since stock prices are difficult to consistently forecast. Recent neuroimaging evidence suggests, however, that activity in brain regions associated with anticipatory affect may not only predict individual choice but also forecast aggregate behavior out-of-sample. Thus, in two experiments, we specifically tested whether anticipatory affective brain activity in healthy humans could forecast aggregate changes in stock prices. Using Functional Magnetic Resonance Imaging (FMRI), we found in a first experiment (n=34, 6 females; 140 trials per subject) that Nucleus Accumbens (NAcc) activity forecast stock price direction, whereas Anterior Insula (AIns) activity forecast stock price inflections. In a second preregistered replication experiment (n=39, 7 females) that included different subjects and stocks, AIns activity still forecast stock price inflections. Importantly, AIns activity forecast stock price movement even when choice behavior and conventional stock indicators did not (e.g., previous stock price movements), and classifier analysis indicated that forecasts based on brain activity should generalize to other markets. By demonstrating that AIns activity might serve as a leading indicator of stock price inflections, these findings imply that neural activity associated with anticipatory affect may extend to forecasting aggregate choice in dynamic and competitive environments such as stock markets.