According to a paper published in the June 29 issue of The Journal of Neuroscience, researchers were able to forecast human actions by analyzing images of test subjects’ brain activity. The results have implications for using neural messages to control prostheses.
Based on active areas in the brain, shown in an fMRI scan, researchers could predict 55 to 65 percent of the time which of three actions a person was going to do, several seconds before it happened. Photograph courtesy of Jason P. Gallivan, University of Western Ontario, London, Canada.
The scientists at the University of Western Ontario (Western), London, Canada, and the University of Oregon (OU), Eugene, used functional magnetic resonance imaging (fMRI) to scan the brains of nine volunteers over the course of the one-year study. The subjects were asked to perform one of three tasks while their brains were scanned. Jason P. Gallivan, a Western doctoral student in the Lab of Neuroimaging of Action and Perception, Centre for Brain and Mind, and his colleagues then watched for changes in activity in different areas of the brain.
According to the study, the researchers found that by using fMRI pattern recognition techniques they were able to decode object-directed grasp and reach movement intentions from human brain signals moments before their initiation. Specifically, the movements involved grasping the top of an object, grasping the bottom of an object, and touching the object. An article in CBC News, a Canadian news-source agency, stated that the researchers could only predict the action four seconds before it happened, with a 55 to 65 percent accuracy rate. Gallivan said he considers the research as exploratory, proof-of-concept work.
“This is a considerable step forward in our understanding of how the human brain plans actions,” Gallivan, the first author on the paper, said in a Western press release.
Previously, scientists had only been able to make similar predictions for animals with electrodes inserted into their brains. “Neuroimaging allows us to look at how action planning unfolds within human brain areas without having to insert electrodes directly into the human brain,” said Jody C. Culham, PhD, associate professor in the Western Department of Psychology, and the paper’s senior author. “This is obviously far less intrusive.”
Gallivan said these finding could also have clinical implications. “A potential implication of the findings is that we may be able to use these same types of intention-related brain signals in order to control prosthetic devices,” he explained.
