Computing How a Brain Sees
Sep 14, 2005 - 2:48:00 AM
|
|
|
How is it that the visual system can recognize certain objects, such as a friend in the crowd, or words on a page, despite the changing environment in which these objects are seen?. Despite the apparent ease in which we see, visual recognition is a very difficult computational problem.
|
By Georgetown University Medical Center,
[RxPG] For his efforts to establish a quantitative framework aimed at describing the complex workings of the human brain, Georgetown University Medical Center neuroscientist Maximilian Riesenhuber, Ph.D., has received a five-year CAREER award from the National Science Foundation (NSF).
The award, one of five given this year by the Behavioral and Cognitive Science Divison of NSF, and the only one in the Cognitive Neuroscience Initiative, is for $742,000. The prestigious NSF CAREER awards are designed to help innovative scientists, early in their profession, develop promising research programs.
Riesenhuber, who came to Georgetown in 2003 after postgraduate (and graduate) study at the Massachusetts Institute of Technology, is the principal investigator for the Laboratory for Computational Cognitive Neuroscience.
As a gateway into understanding information processing in the brain, Riesenhubers research focuses on vision, and specifically on object recognition. Using the NSF grant, he plans to use a variety of methods, including behavioral and imaging studies, to dissect how people recognize things, such as faces, in different scenes.
How is it that the visual system can recognize certain objects, such as a friend in the crowd, or words on a page, despite the changing environment in which these objects are seen? he asks. Despite the apparent ease in which we see, visual recognition is a very difficult computational problem.
Riesenhubers task, therefore, is to use mathematics and computer modeling to make sense of a massive amount of information -- everything from what people say they see to the firing of groups of neurons seen in brain scans -- that goes into visual perception.
Comprehending how a brain sees has a variety of applications, Riesenhuber said. A computational model of vision can be used in artificial intelligence to help machines see, and for applications involving human object recognition tasks, ranging from baggage screening to satellite image analysis.
The model-based approach is also at the core of a grant Riesenhuber has just been awarded by the National Institutes of Health to decipher visual recognition in individuals with autism. In particular, he will study face recognition deficits, which might contribute to the difficulty in social interactions that are a hallmark of the disorder.
"We will use a combination of computational modeling and experiments to go beyond current qualitative theories of how neural information processing differs in typically developing and autistic brains," Riesenhuber said. "This technique might also help us shed light on the neural bases of other disorders that involve a loss or impairment of object recognition abilities, such as dyslexia."
Advertise in this space for $10 per month.
Contact us today.
|
 |
|
Subscribe to Awards & Prizes Newsletter
|
|
|
|
Additional information about the news article
|
About Georgetown University Medical Center
Georgetown University Medical Center is an internationally recognized academic medical center with a three-part mission of research, teaching and patient care (through our partnership with MedStar Health). Our mission is carried out with a strong emphasis on public service and a dedication to the Catholic, Jesuit principle of cura personalisor care of the whole person. The Medical Center includes the School of Medicine and the School of Nursing and Health Studies, both nationally ranked, and the world renowned Lombardi Comprehensive Cancer Center. For more information, go to http://gumc.georgetown.edu.
|
|
Feedback
|
For any corrections of factual information, to contact the editors or to send
any medical news or health news press releases, use
feedback form
|
Top of Page
|
