St. Louis, June 29, 2001 -- With so much of our modern lives dominated by movies and television, it's easy to think of perception as a continuous, unedited, uncut version of the world around us. But new research from Washington University in St. Louis suggests that while we are watching the "movies" of our daily lives, the brain is automatically dividing them into smaller, meaningful units.
Jeffrey M. Zacks, Ph.D., assistant professor of psychology in Arts and Sciences, and colleagues, identified a network of brain areas that is activated during the perception of boundaries between events.
"We found regions of activity in the brain that track the process of identifying parts of continuous events, whether or not people are aware of those parts or even know anything about them," said Zacks. The study is published in the June 2001 issue of Nature Neuroscience.
The researchers observed local brain activity using functional magnetic resonance imaging (fMRI) while participants watched movies of common, everyday goal directed activities, such as making the bed, doing the dishes, or ironing a shirt. In the first scan, participants were instructed to passively watch the movies. These scans revealed that the brain automatically parses continuous events into smaller segments even in the absence of explicit instructions to do so.
"To answer the question about whether event segmentation is an ongoing process, we needed data from when people were not performing any task. So we had them just sit and watch the movies," Zacks said. "The fact that changes in brain activity occurred during the passive viewing of movies indicates that this is how we normally perceive continuous events, as a series of segments rather than a dynamic flow of action."
In two later scans, the same movies were shown again, but participants were instructed to press a hand-held button every time they identified a specific small task -- rinsing, washing, drying, for instance -- that was a necessary component of the larger activity. In one of the later scans, participants were asked to divide the movie into the largest units that were natural and meaningful to them; in the other scan, the smallest units.
These later scans allowed the researchers to look at brain activity during the perception of boundaries between large and small events. "We found local changes in brain activity occurring at the same time people were identifying event boundaries, and throughout this network, changes were larger for large boundaries and smaller for small ones," Zacks said.
Similar changes were observed when participants were passively viewing movies, although to a lesser degree. These findings also confirm previous behavioral work on the hierarchical structure of event segmentation, whereby large meaningful segments correspond to groups of smaller meaningful segments.
The activated parts of the brain include a large bilateral region of the posterior cortex, and a smaller region in the right frontal cortex. The peak of activity in the posterior cortex is in the human MT (or V5) complex, an area responsible for understanding motion and human action.
Activity in the frontal cortex was localized to a part of the frontal eye field (FEF) that is associated with shifting attention. How we form our perceptions has long been a basic question in cognitive science. While much previous research has yielded interesting clues about how we form perceptions in space, much less is known about how we form perceptions across time.
These findings have important practical implications because insights into how the brain processes and perceives individual tasks and activities could be harnessed to suggest more direct and effective approaches to everything from elementary curricula to on-the-job training. "If we understand how the brain and mind chunk activity into tasks, we can design better tools for teaching people new activities," said Zacks.
Zacks can be reached at 314-935-8454 or via e-mail at firstname.lastname@example.org for media assistance, contact Gerry Everding, University Communications: 314-935-6375 or email@example.com.