98-154 (Face Recognition)
Brown University News Bureau

The Brown University News Bureau

Distributed June 22, 1999
Contact: Kristen Cole

Brain region used in face recognition is active in new object recognition
Researchers at Brown University and the Yale Medical School used functional magnetic resonance imaging (fMRI) to demonstrate that the area of the brain active in face recognition is also involved in recognizing new, non-face objects, according to a study in the June issue of Nature Neuroscience.

PROVIDENCE, R.I. -- The brain region critical in face perception is also active when humans become expert in recognizing a set of unknown, novel objects, according to a new study by researchers at Yale University Medical School and Brown University.

The findings indicate the mechanisms responsible for face perception may be a learned skill rather than an innate function of the human brain: nurture, not nature. The primary authors of the study, published in the June issue of Nature Neuroscience, are Isabel Gauthier, a recent Yale Ph.D., and Michael J. Tarr, associate professor of cognitive and linguistic sciences at Brown.

Face recognition generally activates a different area of the brain - the right middle fusiform gyrus - than non-face object recognition, but this study found an expertise effect for non-face objects in the face recognition area of the brain.

The researchers tested five adults using functional magnetic resonance imaging (fMRI), a non-invasive diagnostic procedure that creates a computerized image of the human brain at work. They scanned the five research subjects before, during and after exposing them to a family of novel objects called greebles: once before any exposure; three times during expertise training; and twice after the subjects had become expert at identifying greebles. Six additional adults were scanned only as greeble novices.

Greebles are a homogeneous class of complex three-dimensional objects organized into two categories: gender and family. They were designed in Tarr's laboratory first at Yale and then at the Brown University Department of Cognitive and Linguistic Studies. In this study, the process of training test subjects to be greeble recognition experts took approximately seven hours over at least four days.

When test subjects completed their training as greeble experts, the activity in their right middle fusiform gyrus was similar to activity generated during human face recognition. That brain activity did not occur when greeble novices were tested.

The findings contradict the argument that face recognition is a separate brain mechanism, said Tarr.

"Face recognition is one of the most difficult visual tasks humans perform, because faces are so similar," Tarr said. "Once you understand the mechanisms responsible for face recognition you can work with brain-injured people who have a deficit in that area. The real question now is how this remarkable ability arises."

Future studies should investigate a range of stimuli - both more and less face-like than greebles - and the type of expertise that can generate activity in face-specific brain areas, according to the authors. Their study was supported by grants from the National Science Foundation and the National Institute of Neurological Disorder and Stroke.

######
98-154