New research by Jason Gallivan and Randy Flanagan (Neuroscience/Centre for Neuroscience Studies) reveals that, when humans are presented with two action options, the brain’s motor neurons prepare for both possibilities before deciding which action to take.

This finding supports the idea that the brain represents the world as a series of possible actions and objects to interact with.

“Whether you’re navigating a route to work or browsing produce at the grocery store, our brains are constantly making decisions about movement,” Dr. Gallivan says. “Even outside your conscious awareness, your motor system appears to always be operating in the background, coming up with potential actions. Should I reach for the red or green apple? When should I cross the street?”

Neuroscientists have long debated which comes first – the decision about which target to act on or the movement plan. Motor decisions in the brain happen so quickly that determining the order in which these processes take place is challenging. For this research, Drs. Gallivan and Flanagan devised a task that separated visual targets from the movement needed to reach them.

Sixteen volunteers were asked to steer a cursor towards one of two targets, but the catch was they had to start the movement before finding out which target they’d have to pick.

“When you’re forced to launch an action without knowing which target is going to be selected, people simply launch actions that are right down the middle, between the targets,” Dr. Gallivan says. “This minimizes the corrections needed to reach each target. However, we don’t know whether this reflects a deliberate strategy to aim in between the target locations or a byproduct of two possible movements having being prepared in parallel.”

In a twist, there was a hidden feature in the task. At first, the position of the cursor matched the position of the hand exactly, but with each repetition of the task, the cursor slipped a little bit more out of sync with the hand.  Because the change was so gradual and because the volunteers couldn’t see their hand, people were unaware that they gradually compensated for the hand-cursor mismatch by altering their hand movement.

When the researchers analyzed the data, they found the volunteers’ hand movements were the average of the movement paths needed to reach the two potential targets, and not an average of the visual target directions. This suggests that people were unconsciously preparing for the two potential movements prior to having selected the final target.

“Our work portends itself to the design of more intelligent robotics,” Dr. Gallivan says about the results. “Whereas we have been able to design computer programs that can beat the grandmasters at chess, we have yet to design a robot that can interact with the world in a way that matches the manual dexterity of a five-year-old child.”

The research was recently published in Cell Reports.