Walk This Way: Mason Researcher Looks at Human Movement and Perception
Posted: October 29, 2007 at 1:00 am, Last Updated: November 30, -0001 at 12:00 am
Even from a distance, most people are able to recognize close friends and family members by the way they walk. Psychology professor James Thompson is studying why.
Photo by Evan Cantwell
You can probably pick your mother, your brother, your husband or best friend out of a crowd from a distance. From across campus, with just the briefest glance, you can recognize them walking toward you.
You might also have had the sensation of seeing a stranger who reminds you of someone you know but you’re not quite able to pinpoint who it is or why they remind you of someone else.
Photo by Evan Cantwell
For Jim Thompson, assistant professor of psychology, the reason behind these everyday recognitions of movement is his puzzle to solve. Thompson studies how our brains recognize people based on their walk and other movements. Heading the Perception and Action Neuroscience Group (PANG) at Mason, Thompson and his students are focused on examining how we recognize human movement and make sense of other peoples’ actions.
“The ability to see how other people move is essential for many aspects of daily life – from things as simple as avoiding collisions to detecting suspicious behavior or recognizing someone else’s emotions,” says Thompson.
Thompson says we have an “embodied understanding” that allows us to quickly recognize people and process and understand their movements. The way in which we understand other people’s actions is often based on our own actions. Our brain sets “goals” for other people’s actions that meet our own.
For example, when you see someone pick up a pen and start writing with it, you can match that action to similar actions that you have performed and make a guess about what the person’s goal is. However, if you see someone pick up a pen and start eating it, you have a very different reaction.
“Your brain processes that action in a very different way and then must reevaluate the goals for that action,” says Thompson.
Observing Human Actions
In a 2005 study done by Maggie Shiffrar and her colleagues at Rutgers University, participants were asked to identify people making various movements with point-lights attached only to their major joints. Most of the respondents were more quickly able to identify themselves moving than they were able to identify close friends. This suggests motor experience influences the visual analysis of action, that is, we can better recognize our own movements — even though we rarely see ourselves move except in reflection — than the movements of friends or strangers.
Thompson’s research builds on the research done by Shiffrar by examining which areas of the brain become active when watching someone else walk. He uses functional magnetic resonance imaging (fMRI) to pinpoint the areas of the brain that are active when we watch someone move and combines that technology with electroencephalograpy (EEG), the neurophysiologic measurement of the electrical activity of the brain. Once he has a precise timing of the reaction of the brain’s activity, Thompson can get a very distinct picture of the brain’s reaction to other people’s movements.
This image shows an individual’s brain activation while viewing human movement. The warm colors represent functional magnetic resonance imaging (fMRI) increases to viewing human actions, overlaid on a high-resolution structural MRI scan. These data were gathered at Mason’s MRI facility at the Krasnow Institute for Advanced Study.
Image courtesy of James Thompson
His studies have proven that visual and motor parts of our brain “light up” when we watch someone walking. There is growing evidence that we use motor regions — parts of the brain that we use when moving our arms or legs — when merely looking at other people move.
“I am endlessly fascinated by how the brain is able to solve complex puzzles like recognizing human motion,” says Thompson. “At the moment, computer science is making great progress in designing hugely complex computer programs to do tasks like this that we, as humans, can do very easily.”
You Seem Very Familiar…
Another area of research in PANG looks at the question of how do we recognize people just from the way they walk, dance or sit down? What process does our brain go through? How does our brain categorize different walks and rank them?
“There are many subtle differences in the way people walk,” says Thompson. “Some people swing their arms more than others, others sway their hips more.” It might be these small differences working together as a pattern that allow us to tell one person from another.
“This research is one part of a much bigger question of how we recognize people and interpret what they are doing, something that is very important for all aspects of social interaction,” says Thompson.
His research can help in specialized settings such as surveillance and in conditions in which human movement recognition may be impaired. Thompson, along with Raja Parasuraman from Mason’s Department of Psychology and Scott Grafton from the University of California at Santa Barbara, received a one-year grant for $370,000 from the Army Research Laboratories to study the brain basis of action recognition.
“In combat situations, you need to rapidly recognize if someone is a friend or a foe,” says Thompson. “Understanding how we process the actions of others might assist in developing training so that soldiers can know where to focus and what to look for.”