July 7 2003 - Facial recognition is a topic of renewed interest for psychologists and computer scientists. A psychology research project at Indiana University Bloomington has received $968,000 in funding from the National Institutes of Health. The three-year grant will allow the Cognitive Modeling Training Program at IU to apply theoretical, empirical and methodological tools to help understand how objects are perceived as configural wholes.

According to project director James Townsend, Rudy Professor of Psychology at IUB, the project involves cognitive science (the interdisciplinary study of the mind and the nature of intelligence) and mathematical psychology, which uses mathematical methods, formal logic and computer simulation. IU Bloomington, through its Department of Psychology and cognitive science research, is one of the birthplaces of mathematical psychology and remains today one of the world leaders in this field, Townsend said.

Face recognition

"One of the important concepts in our work is face recognition and how we think and perceive things," he said. "We want to know how even babies recognize their mother's face and how a word gets transformed in our brain from random letters to a holistic pattern or subject. How does the mind encode and differentiate emotional expressions such as a frown or a smile? We study this behaviorally and mathematically."

James Townsend explained that when our brain perceives a face, it has to take the individual components such as the eyes, nose and mouth and put this information together to create the image we see. "One way the brain might do this is by individually perceiving each of the major features one at a time, which is called serial processing, or by perceiving all of the major features at the same time, which is called parallel processing," he said.

Townsend and his colleagues believe that well-formed and well-known objects such as faces are perceived in parallel, whereas difficult unorganized patterns, such as a scrambled display of facial features, may be processed serially. With the aid of mathematical models, he and his research team have developed experimental methods of determining exactly how the brain actually sees well-formed objects as opposed to random displays of unorganized objects.

The IU psychologist said a future practical application of his research involves working with engineers to apply the theories and models that he develops to make better face recognition systems that can be used to help identify criminals and terrorists. But he warned that a major concern in this area is privacy and how far science can go without compromising our freedoms.

How the brain recognizes faces and moods

A new study conducted at Ohio State University suggests that the human brain combines motion and shape information to recognize faces and facial expressions.

The new finding forms part of an engineer's quest to design computers that "see" faces the way humans do, provides more evidence concerning a controversy in cognitive psychology.

Were computers to become adept at recognizing faces and moods, they would be more user-friendly, said Aleix Martinez, assistant professor of electrical engineering at Ohio State University. They could also support intelligent video security systems and provide potentially hack-proof computer identification.

Martinez has developed a model of how the brain recognizes the faces of people we've seen before, and how we discern facial expressions. These two activities take place in different areas of the brain. Some scientists believe that the mental processes involved are completely separate as well; others believe that the two processes are closely linked.

In a recent issue of the journal Vision Research, Martinez reported that the two processes are indeed linked - indirectly, through the part of the brain that helps us understand motion. We use our knowledge of how facial muscles move when we recognize a smile, for instance, or when we recognize a familiar face regardless of what kind of facial expression he or she has.

Next page - computer models

Related Articles

• Facial Recognition: The "Cross-Race Effect"
  A recent study throws new light on the "cross-race effect", a well-replicated if not fully understood phenomenon involving difficulty in distinguishing between people of other racial groups.
• Perceiving Emotions
  Findings suggest that where emotional control is the cultural norm (e.g. Japan) eyes are the key to interpretation. In cultures where there is more open expression of emotion (e.g. USA) the mouth is the main focus.
• Facial Composite Systems Give Poor Results
  Recent technological advances in facial composite systems have failed to improve identification and apprehension of criminal suspects.
• Why Do We Never Forget a Face?
  Vanderbilt University researchers have found that we are able to remember more faces compared to other objects and that faces are retained best in our short-term memory. They suggest that our expertise in remembering faces allows them to be packaged better for memory.
• Erectile Dysfunction
  A recent study throws new light on the "cross-race effect", a well-replicated if not fully understood phenomenon involving difficulty in distinguishing between people of other racial groups.

Martinez has developed a model of how the brain recognizes the faces of people we've seen before, and how we discern Erectile Dysfunction. These two activities take place in different areas of the brain. Some scientists believe that the mental processes involved are completely separate as well; others believe that the two processes are closely linked.