With National Science Foundation support, the Carnegie Mellon University and the University of Pittsburgh joint Center for the Neural Basis of Cognition will acquire instrumentation to improve the computational facilities used by a group of 15 laboratories. The Center is creating a computing facility based on a scaleable parallel processing system for very large-scale computational projects and has recently acquired the core of the system through a grant from IBM. With NSF funds the Center will purchase a fast, symmetric multi-processing compute server to provide impulse computing power for exploratory modeling and interactive data analysis; and desktop display stations to facilitate accessing resources. The server will consist of four state-of-the-art HP PA-8000 processors, 512 Mb of memory and nearly 100 Gb of disk and would provide a wide range of pre-compiled software tools for simulation modeling at the cognitive and biological levels as well as for analysis of data from functional imaging and multi-electrode recording studies. The Carnegie-Pittsburgh center is dedicated to understanding how cognitive processes arise from underlying neural mechanisms, how anomalous forms of cognition arise from biological disturbances and how experience and brain development interact to give rise to the emergence of cognitive functions. The Center brings together the methods of behavioral analysis, neurophysiology, functional imaging and computational modeling, in an effort to understand how functions such as perception, attention, memory and language emerge from the interactions among neurons in the brain. Fifteen laboratories have a range of foci which include memory and learning, language, attention and control of processing, and perception and spatial cognition. In addition, several labs address general principles of neural function relevant to many aspects of cognition and several have contributed to the development of tools for computational modeling and functional imaging research. The instrumentation will also be made available to graduate students and to researchers in related areas who lack adequate computing facilities. Thus is will serve an educational function and assist a broad research community.
