6
The Biological-Electronic Interface
Two of the great revolutions of the 20th century have occurred in molecular and organismic biology, and in information processing. Although it is a central part of both biology and computation and communications, information has very different currencies in biology and in electronic computation. In biology, information is stored in distributed networks (neural networks) in the form of protein in junction regions between cells; the “current” is a combination of ionic movement and the movement of molecular neurotransmitters; memory is still undefined biochemically. In electronic computers, information is stored as electrons in capacitors, the current is electrons moving, and memory is ranks of addressable capacitors or arrays of polarized magnetic domains. Both biological and electronic information processors are powerful in their own sphere: biological systems for self-assembly, adaptability, and pattern recognition; electronic systems for speed, stability, and manufacturability. At present, there are no systems that fuse biological and electronic information systems.
The ONR has led in the fundamental science of information processing and has strong programs in biology and biotechnology. There is an opportunity for ONR to form a program to attack one of the largest opportunities in basic science: that is, fusing biological and electronic information processing. The benefits of a program that would enable electronic and biological information processing systems to talk to one another would range from basic scientific understanding of the architectures of biological systems (neural networks represent an example that has already proved important in computation), to neural and perceptual prostheses, to very long range visions of systems that are partly biological and partly electronic and are able to combine the most characteristic features of each.
Important components of such a program, as it might now be imagined, would include:
-
New science to control the interface between man-made “stimulators/receivers” and nerve cells. At present, nerves are grown on electrodes and stimulated electrically. It is not clear that this strategy is the best one, and innovative thinking is required about ways of building an enduring and functional biological-electronic interface. It does seem clear that chemistry is the enabling science that will be key to understanding and utilizing the interface. This work would combine interface science, attached cell cultures, organ cultures, microfabrication, and cellular neurobiology.
-
New methods to understand the nature of information processing in biological systems, and to model it in electronic systems. Nature has a remarkable range of solutions for problems in perception and information processing and has robust capabilities—ranging from very fast pattern recognition to self-awareness—that can now be found nowhere in electronic systems. Understanding and modeling these systems will require innovative work in theory and modeling, new analytical systems at all scales from molecular to anatomical, the construction of realistic models for complex biological systems, the understanding of control and stability in these systems, and the design and prototyping of model systems.
-
Engineering implementation of basic science in the form of working prototypes. The design of true hybrid systems—that is, systems that combine biological and computational components—represents an enormous challenge. Consider, for example, the difficulties of simply taking the compound eye of an insect and using it stably with a computer and information processor as a sensor. The reverse process—using a computer to generate signals that would be interpreted by the brain as “seeing”—is sufficiently difficult that there is no real single point of entry into the problem. Any attack on this problem will require the combined effort of virtually all the disciplines touching on the field: electrical engineering, biology, materials science, information science, and microfabrication.
ONR is in a unique position to lead research in this area. The diversity of experience and knowledge represented by the ONR scientific officers would allow the establishment of a focused research program in this exciting area and would minimize the potential for wasteful diversions.