Biotics comprises two complementary areas of application: that of analog signals (in this case, bioelectronics) and that of digital signals (molecular electronics). The construction of a "biocomputer" based on circuits and memory from DNA or molecular electronics and using materials compatible with living systems, is part of biotics. (Aldeman 1994,  Kolata 1995, Kari 1997). The field emerged from recent advances in biology, solid-state physics, organic chemistry, micro-electronics, robotics, and nanotechnology. Today it constitutes a new area of research with many applications.

Molecular electronic components are currently considered the potential successors to semiconductors. These synthetic components offer many advantages over traditional semiconductors: three-dimensional assembly, synthetic materials that allow the custom design of properties, miniaturization approaching that of biological structures, and possibilities for interfacing with living systems. (Reed  1999, Joachim 2000, Tour 2000).

New interfaces between the human brain and computers
Direct neurons to machines interfaces have been developed during the last years. Boris Rubinsky at Caltech has developed a “biotransistor” made of living cells interconnected with a microprocessor. The cells acts as diodes. (Rubinsky 1999). William L. Ditto, a physicist at the Georgia Institute of Technology working with a group at the University of Bordeaux in France, has developed hybrid computers that mate living neurons with silicon circuits. He has called this new field “neurosilicon computers.” In 1999, he was able to do arithmetic with two large neurons from leeches, joined together and linked to a personal computer. (Spano and Ditto, 1999)

Another step on the road to creating biocomputers was achieved by Jerry Pine, a biophysicist at the California Institute of Technology in Pasadena. He was able to “grow” microcircuits made of living neurons on top of an array of electrodes. He calls the device the “Neurochip.” By assigning a specific place to each neuron, it is possible to “listen” to their chatter and develop reproducible logic gates out of combinations of neurons.(Regehr and  Pine 1988, Maher and Pine in press, Pine et coll 1996).

A similar type of research has been undertaken by Keiichi Torimitsu at the NTT’s Biosciences Research Group in Atsugi, Japan. (Niwa and Torimitsu, 1998 His group is trying to develop an effective interface between computers and the brain. To test this possibility, his laboratory sent electronic signals to slices of neuronal tissue placed close to tiny electrodes and researchers monitored the electronic current naturally generated by the neurons when they communicated with each other. (Torimitsu 1998).

More recently, Miguel Nicolelis of the Duke University Medical Center, has trained two owl monkeys to control a robotic arm through brain signals. The arm was placed at MIT’s lab for Human and Machine Haptics and controlled by the monkeys through an Internet interface. (Nicolelis 2000)

Biological evolution and the Internet as massive parallel multiprocessors
The new symbiotic interfaces between man, computer and networks, creates a massive parallel multiprocessor. Biological evolution performs like such parallel multiprocessor. The basis of biological evolution is the three-part process of mutation, competition, and selection. Random variations occur in the programming of living things (DNA). This results in new species that are more or less suited to the environment in which the species are in competition. The fittest survive, are selected — or rather self-select — and transmit to their descendants the genetic code for survival and competitiveness, the new mutant genes. This process takes place in parallel within the DNA of billions of individuals in competition for limited resources. Biological evolution, is therefore comparable to a huge parallel multiprocessor that seeks solutions to problems by trying out potential solutions and storing those that work in memory. This is how the diversity of the living world, biodiversity, is created. (de 1995,  2000)

The autocatalytic development of the Internet is an illustration of a coevolutionary process of order emerging from chaos. Millions of agents acting in parallel according to simple rules also form a gigantic multiprocessor that can collectively find solutions to complex problems and adapt to the evolution of its informational ecosystem. As a result of these rules and emergent properties, the Internet has become an increasingly intelligent planetary metacomputer. It processes data in parallel, combining the actions of millions of agents testing procedures and programs in real time in a competitive environment — a process that is not unlike Darwinian biological evolution. We can therefore expect the Internet to select increasingly powerful solutions in electronic communications and software applications. Intelligent agents manage interfaces by interconnecting all the existing networks, allowing people to access information and act in real time, as do the neurons of the brain. This new planetary neural hypernetwork functions chaotically, fluidly, and in a way that is constantly reconfigurable, in response to decisions made in parallel by hundreds of millions of interacting human agents and virtual robots. In this, it resembles the immune system, the hormonal system, and the nervous system, three interconnected networks that determine an organism’s psycho-neuro-immunological behavior.

Selective stabilization and reconfiguration of Internet links and nodes
In 1949, in his book The Organization of Behaviour, Donald O. Hebb, a neurophysiologist at McGill University, in Montreal, proposed a revolutionary new theory of psychological behavior. (Hebb 1949). According to this theory, the brain constantly reconfigures the synapses that transmit nerve impulses. Through the chemical action of activator or inhibitor hormones, the synapses are reprogrammed as a result of various stimuli. Through the successive stimulation of neural connections and pathways, whole areas made up of thousands of neurons are activated and connect so as to form subsets that store information through the reinforcement of impressions (shapes, colors, sounds, words). These subsets constitute dynamic networks of neuronal interactions, the brain’s building blocks of information.

I propose to look at the formation and functioning of the Global Brain, (that I call “the Cybiont”), in a similar way. Human beings, multiple agents in chaotic interaction, are the neurons of the hypernetwork. The links among them, occurring through computers (and even more directly through biotic interfaces), are giving rise to a conscious representation of the “mental” functioning of the Cybiont, a global consciousness that is reflected in the introsphere. (de Rosnay 1995, 2000). These links are reversible, and they can be reinforced or inhibited. Autocatalytic processes take place, leading to new concepts, solutions, or ideas. The Internet today abounds with examples of such processes. (Heylighen 1996)

Jean-Pierre Changeux, of the Institut Pasteur and his collaborators have proposed a model of epigenesis of neural networks by selective stabilisation of synapses and analysed in these terms the molecular mechanisms involved in the regulation of acetylcholine receptor genes expression during the development of the motor endplate. (Changeux 1985, Kerszberg and Changeux 1992). In particular, they have identified DNA regulatory elements, as first/second messengers, specifically involved in the regulation of acetylcholine receptor genes transcription by electrical activity in extra junctional areas, and by "trophic" factors in the subneural domain. These issues are of relevance for the understanding of long term synaptic plasticity.

I propose that the selective stabilization of Internet node follows an analogous principle through HTML links, bookmarks, address books, Web sites, creating a situation of intercommutability and increasing the complexity of the network. New properties will emerge from such highly complex system.

Autocatalytic processes, self selection and emergence of new properties
Emergence, mutation, and breakthroughs can be observed in the light of rapid accelerations resulting from sudden phase transitions. These phenomena are typical of the Internet’s fast development, and can be seen as autocatalytic systems creating dense “time bubbles” and fostering the emergence of new properties through rapid phase transitions. To illustrate this type of self-selection, I propose to adapt to the Internet the random graph model used by Stuart Kauffman to outline the role of collectively autocatalytic molecular systems in the origins of life. (Kauffman 1995). His model is based on the interconnection of many buttons using threads. As Stuart Kauffman puts it, “When there are very few threads compared to the number of buttons, most buttons will be unconnected. But as the ratio of threads to buttons increases, small connected clusters begin to form. As the ratio of threads to buttons continues to increase, the size of these clusters of buttons tends to grow.” A phase transition suddenly occurs when the ratio of threads to buttons reaches 0.5, and a giant cluster is formed. The rate of growth of the giant cluster then slows down as the number of isolated buttons and small clusters decreases. This is represented by the top of the S-shaped curve.

I propose to replace the buttons with web sites (nodes) and the threads with Internet links (edges). Let’s imagine millions of web sites and millions of links. Beyond a given ratio of links to web sites (0.5?), a phase transition must occur. With 400 million users, 170 million host computers, and an average of 50 links per site (bookmarks and email addresses), new properties will certainly emerge. What about with 2 billion users, 800 million host computers, and 500 links per site? With such a giant electronic cluster of interconnected brains and machines, what will these properties look like? Probably a new form of macrolife becoming progressively conscious of its own existence and self-maintenance.
 

Bibliography

Aldeman, Leonard , « Molecular Computation of Solutions of Combinatorial Problems », Science, 266 : 1021-1024, November 11, 1994.

Changeux, Jean-Pierre, « Neuronal Man: The Biology of Mind », Oxford University Press, Oxford, 1985

de Rosnay, Joël, " Les biotransistors: la microélectronique du XXIème siècle ", La Recherche, n° 124, Vol 12, Juillet-Août 1981, pp. 870-872.

de Rosnay, Joël, " La biotique : vers l’ordinateur biologique ? ", L’Expansion, 1er-21 mai 1981, pp.149-150.

de Rosnay, Joël, " From Biotechnology to Biotics : the Engineering of Molecular Machines ", in : Biotechnology : Applications and Research. ed by Paul N. Cheremisinoff and Robert P. Ouellette, chapter 1, pp. 1-8 Lancaster: Technomic Publishing Co, inc. 1985

de Rosnay, Joël, " Molecular Information Processing and Molecular Electronic Devices ", Fifth International Conference on Langmuir-Blodgett Films, Août 1991, Cité des Sciences et de l’Industrie, La Villette, Paris..

de Rosnay, Joël, " The Symbiotic Man: A New Understanding of the Organization of Life and a Vision of the Future", McGraw-Hill Professional Publishing, March 31, 2000.
“L’Homme Symbiotique, regards sur le troisième millénaire”, Editions du Seuil, 1995

Heylighen, Francis and Bollen, Johan, "The World-Wide Web as a Super-Brain: from metaphor to model" in: R. Trappl (ed.) (1996): Cybernetics and Systems '96 (Austrian Society for Cybernetic Studies), p. 917.

Hebb, D.O., « The Organization of Behaviour », New York, Wiley, 1949.

Joachim, C., Gimzewski, J.K., Aviram, A., « Electronics using Hybrid-molecular and Mono-molecular Devices », Nature, volume 408, pages 541 à 548, November 30, 2000.

Kari, L., “DNA computing: the arrival of biological mathematics”. The mathematical intelligencer 19:9-22, 1997

Kauffman, Stuart A., “At Home in the Universe : The Search for Laws of Self-Organization and Complexity”, Oxford Univ Press; October 1995.

Kolata, Gina. “A vat of DNA may become the computer of the future”. N.Y. Times News Service, 1995

Kerszberg, Michel, Stanislas Dehaene, and Jean-Pierre Changeux, “Stabilization of complex input-output functions in neural clusters formed by synapse selection”, Neural Networks, volume 5 (1992), number 3 pp. 403-414.

Maher, M. P., Pine, J., Wright, J. and Tai, Y.-C.,  "The Neurochip: A new multielectrode device for stimulating and recording from cultured neurons," J. Neurosci. Meth., in press.
http://www.its.caltech.edu/~pinelab/mike.html#neurochip 
http://pr.caltech.edu/media/Press_Releases/PR11839.html

Nicolelis, Miguel A. L. et all, "Real-time prediction of hand trajectory by ensembles of cortical neurons in primates", Nature, volume 408, pages 361 to 365, November 16, 2000

Niwa, Osamu and Torimitsu, Keiichi,  “Real-time Measurement of Neurotransmitters Released from Cultured Nerve Cells with Online Enzyme Sensors”, NTT Basic Research Laboratories, 1998

Pine, J., Maher, M., Potter, S.M., Tai, Y.-C., Tatic-Lucic, S. and Wright, J. (1996). “A cultured neuron probe”. Proceedings, IEEE-EMBS Annual Meeting, Amsterdam 10/31/96. 

Reed, Mark A. et coll « Conductance of a Molecular Junction », Science, Vol. 278, pages 252–254; 10 Octobre, 1997

Regehr, W.G., Pine, J., and Rutledge, D.B., "A Chronic In-vitro Neuron-microdevice Connection,", IEEE Trans. Biomed. Eng'g, 35, 1023 (1988).

Rubinsky, Boris and Huang, Yong « A Microfabricated Chip for the Study of Cell Electroporation », Biomedical Engineering Laboratory, Department of Mechanical Engineering, University of California, Berkeley CA 94720, February 1999.

Spano, M.L. and Ditto, W.L., “Chaos control in biological systems”, in : Handbook of Chaos Control, pages 427 - 456, 1999). 
http://www.businessweek.com/1999/99_25/b3634137.htm, http://gtalumni.org/news/magazine/sum00/profile.html

Torimitsu, Keiichi et coll., “Application of Microfabricated Biosensor Chip for Neuroscience”,  Electrochemistry, Vol. 68. No. 04, 2000, p. 284

Tour, James M. and Reed, Mark A. « Computing with Molecules », Scientific American,  June 2000