News Release

People can learn motor skills by watching

Peer-Reviewed Publication

Cell Press

It's widely accepted that people watching an expert golfer or carpenter can learn the procedural steps to a better golf swing or building a deck. However, researchers Andrew A.G. Mattar (presently at McGill University) and Paul L. Gribble of the University of Western Ontario have developed startling evidence that people can unconsciously learn complex motor behaviors by watching such performances.

Their experiments indicate that subjects who watched a video depicting a person learning a motor skill would engage their neural systems for movement planning and control, and that the subjects would construct a neural representation of the mechanical requirements of the task.

"A powerful new idea in neuroscience links motor control with action observation," wrote Mattar and Gribble. "When we observe the actions of others, we activate the same neural circuitry responsible for planning and executing our own actions." In their experiments, the researchers tested "the intriguing possibility that such a system linking observation and action could facilitate motor learning."

In the experiments, the researchers asked subjects to move a handle that was part of a robotic device to guide an on-screen cursor to targets. They then divided the subjects into groups who watched videos of a person learning to operate the device when the handle was subject to either a clockwise or counterclockwise force field. A third group was simply asked to rest and watch nothing. The subjects were not told about the forces or their roles in the experiment.

The researchers found that, compared to subjects who watched nothing, those who observed the video of the "clockwise force field" learning performed significantly better when they themselves had to operate the device under a clockwise force. Importantly, those who had watched the "counterclockwise force field" video performed worse than those who watched nothing.

The researchers tested whether the video watchers were covertly practicing, by recording subjects' arm muscle activity during video watching. They found no such muscle activity when subjects watched the videos.

Also, the researchers tested whether the learning process was conscious, by distracting subjects during video observation--asking them to do an arithmetic addition task. Mattar and Gribble found that the distraction did not change the motor learning benefit from watching the video. And, when interviewed, the subjects did not show any conscious awareness of how the forces that they actually used compared to those that they saw on the video.

However, the beneficial effects of observing were significantly reduced when the subjects were asked to perform unrelated arm movements during observation.

"These findings broaden the scope of theories linking observation and action by demonstrating that by watching another individual learning to move, observers can learn not only what movements to make, but how to make them as well," wrote the researchers.

"Here we have shown that by observing another individual learning to move accurately in a novel mechanical environment, observers move more accurately themselves," concluded the researchers. "Subjects can acquire neural representations of novel force environments on the basis of visual information."

"The finding that subjects can learn something useful about novel force environments on the basis of observation is remarkable, given the complex relationship between movement kinematics and associated time-varying neural control signals to muscles," they wrote.


Andrew A.G. Mattar and Paul L. Gribble: "Motor Learning by Observing"

Publishing in Neuron, Volume 46, Number 1, April 7, 2005, pages 153–160.

The researchers include Andrew A.G. Mattar of the University of Western Ontario (presently at McGill University) and Paul L. Gribble of the University of Western Ontario. The research was supported by CIHR (Canada).

Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.