Scientists can accurately forecast stock prices — using human brain activity

STANFORD, Calif. — Scientists say activity in regions of the brain associated with seeking reward and taking risks can predict price changes in the stock market more accurately than previous trends or investments. Activity in the brain has already been used to predict what videos will go viral and whether crowdfunding campaigns will succeed. Now, researchers at Stanford University say “neuro-forecasting” can also be used to beat the stock market.

It’s a superpower every investor and financial analyst would love to have: the ability to accurately predict stock trends. Of course, forecasting stock prices is very challenging as it depends on other people’s investment choices. Studies have shown that brain activity can be used to determine what choices people are likely to make as a group.

Whole brain confirmation that activity in predicted regions forecasts stock price direction and inflection. Top, Stock price direction: NAcc activity forecast stock price direction in experiment 1 (middle), but not experiment 2 (right). Bottom, Stock price inflection: AIns activity forecast stock price inflection in experiments 1 (middle) and 2 (right).

“Successful investing is challenging, since stock prices are difficult to consistently forecast. However, new evidence suggests that anticipatory affective brain activity may not only predict individual choice, but also may forecast aggregate choice,” the authors write.

An experiment was carried out by the researchers involving 90 volunteers with little or no financial investing experience. Participants were asked to examine real stock price trends from 2015 and decide whether they wanted to buy or sell the displayed stocks. For each stock, volunteers were given $10, after which they made 10 consecutive investment choices.

If volunteers invested and the stock price went up, their balance was increased by $1.00. While they were deciding, the researchers scanned their brains using a functional magnetic resonance imaging machine (fMRI). In particular, they measured activity in the nucleus accumbens and anterior insula, areas of the brain involved in seeking reward and avoiding risk.

Based on their results, they were able to forecast how a stock would behave the next day. Researchers found if activity increased in the nucleus accumbens, the stock price was going to rise. Alternatively, if activity increased in the anterior insula, it was going to flip or change direction.

“We found that nucleus accumbens activity forecasts stock price direction, whereas anterior insula activity forecasts stock price inflections. Importantly, group choice behavior could not forecast stock prices, implying that the findings could not be attributed to learning over time or to correlated stock price histories,“ the study explains.

Researchers also say that previous stock market trends and the participants’ own investing choices did not predict price changes.

“These findings challenge traditional assumptions of market efficiency by implying that neuro-imaging data might reveal hidden information capable of foreshadowing stock price dynamics. Additionally, they challenge traditional theoretical accounts which imply that elements of choice cannot inform financial forecasts,” said Dr. Stallen.

The findings are published in the journal Neuroscience.

SWNS writer Tom Campbell contributed to this report.