|
Interactive Computation at the Bio/Info/Micro Interface Coupling of Brain with Microstructured Electronic/Optoelectronic Arrays Interactive Computation at the Bio/Info/Micro Interface Statement Of Objectives We propose to embark on a new interdisciplinary approach to the study and use of complex, distributed information processes which exploits the interaction between a biological processor (mammalian brain) and a man-made massively parallel network of nanoelectronic and microphotonic sensor/transducer arrays. We propose to explore a set of specific physical arrangements where signals from each subsystem influence each other to create a coupled system, quite literally at the bio/info/micro interface. Through the implementation of these physical vehicles and the use of the specific properties and advantages of both types of systems, neuronal firing patterns, correlation, cross-correlation and coordinated activities, both in the absence of stimulation and under electrical and/or photonic stimulation, will be examined, recorded, and studied on a unprecedentedly large scale with an unprecedentedly fine resolution. New pathways for extracting computational functions from collective behavior of massively coupled nanoelectronic/photonic systems will be explored by experimenting with the new degree of freedom of physically mapping the neuronal connectivity onto the array of nanoelements. This project will promote further understanding of the neural and synaptic organization and information processing capabilities of human brain by developing new means of acquiring insight to the long range, correlated information processing which underlies the remarkable capabilities of the brain, such as those encountered in sensory and learning processes. It should give guidance to the design of future man-made computers using nanometer-scale components while taking cues from the functional architectures underlying correlated activity in cerebral cortex. Through distributed measurements of the dynamics of neural networks in an operating biological system, or even by projecting its connectivity directly onto an otherwise homogeneous but excitable array of nanoelements, we hope to learn how the neuronal connectivity of cerebral cortex enables the long-range coordinated activities (computation?) of the cortical networks to seek clues for developing architecture of future massively parallel computers. As the basic nanoelectronic system component, we will explore arrays of carbon nanotubes for their significant potential utility in providing massively parallel interfaces to communicate with neurons, cells and dendrites in cerebral cortex. As the basic microphotonic system component, we will pursue the development of highly compact arrays of semiconductor light emitters (especially those at blue and ultraviolet wavelengths) and photodetectors to implement optoelectronic sensors/transducers capable of micrometer scale spatial resolution. The nanoelectronics and photonic elements will integrate provide a powerful subsystem on the manmade side of the bio/info/micro interface. In neuroscience, advances in understanding of correlated behavior of neurons have been made from measurements using relatively low spatial resolution silicon needle or microwire probes. It is possible to record ~50-100 neurons; far fewer than required to explore large scale interactions. Building on the present base of understanding, we will be able to exploit the new sensor array technologies with unprecedentedly high spatial resolution and range to examine correlations in activity among ensembles of coupled neurons in a large network. The potential technical advantages and interdisciplinary synergies realized from this collaborative effort are expected to be profound. Since the biological and manmade systems are both, in some sense experimental and empirical, this project will require intense collaboration and sharing of techniques, insights, and methodologies among the team members in order to create a real basis for success in the total program objectives. |
|
|
|
Copyright © 2003, Center for Advanced Materials Research |