Stahl67) bmp

THE ROLE OF MODELS IN THEORETICAL BIOLOGY 193

VI. Cybernetic Models

A large class of models shall be lumped under the concept of cybemetics. This includes simulation of physiological control mechanisms, neuron networks, brain action, thought algorithms, biological automata, and self-organizing systems. It also includes certain purely abstract mathe-matical models of biological systems. The term cybemetics was men-tioned by Plato and used by Ampere in 1843—in Greek it means “steers-manship.” It was greatly popularized by Wiener (1961) who defined it as the science of “ communication and control in machines and organ-isms.” Other useful definitions include: “the mathematical science of control” or “science of optimal control.” Biological cybemetics con-centrates in a major way on construction of models, commonly by use of digital computers. This section can only provide representative refer-ences conceming cybemetics models.

A number of books, reviews, and symposia volumes are available in this field. The basie work on cybemetics by Wiener (1961) was cited above. Wiener and Schade (1964) edited a collection dealing with ap-plication of cybemetics to the endocrine system, sensory function, celi regulation, and simulation of human thought. Ashby is author of two significant works in biological cybemetics: “Design for a Brain” (1960) and “An Introduction to Cybemetics” (1956), both of which deal very specifically with simulation of biological control mechanisms. Such modełing has, in generał, been a prime goal of cybemetics, whereas bionics stresses construction of physical analogs which function in the manner of efficient biological systems. Grodins (1963) is the author of a book on biological control mechanisms, while Sayre and Crosson (1963) have edited a collection called “The Modełing of Mind: Computers and Intelligence.”

Rosenblith, an early co-worker with Wiener in the field of biological cybernetic modełing, is editor of a symposium (Rosenblith, 1962) on biophysical models, including those concerned with physical (electrical) neuron models, artificial neuron networks, self-organizing systems, information flow, etc. Another symposium volume (Stark et al., 1965) deals with the simulation of the fuli rangę of brain activities, starting with molecular celi models based on algorithm theory to purely pheno-menological simulations of human symbol manipulative processes. The “General Principles of System Analysis” by Quastler (1965) places a strong emphasis on simulation. George (1960) deals specifically with “Models in Cybemetics.” Information theory is used as a basie tool in