An embodiment of some vertebrate command and control principles

Abstract

Throughout the life of the vertebrates, the core of the central nervous system, the reticular formation, has retained the power to commit the whole animal to one mode of behavior rather than another. Its anatomy, or wiring diagram, is fairly well known, but to date no theory of its circuit action has been proposed that could possibly account for its known performance. Its basic structure is that of a string of similar modules, wide but shallow in computation everywhere, and connected not merely from module to adjacent module, but by long jumpers between distant modules. Analysis of its circuit actions heretofore proposed in terms of finite automata or coupled nonlinear oscillators lias failed. We propose a set of nonlinear, probabilistic, hybrid computer concepts as guidelines for specifying the operational schemata of the above modules. Using the smallest numbers and greatest simplifications possible, we arrive at a reticular formation model consisting of 12 anastomotically coupled modules stacked in columnar array. A simulation test of its behavior shows that despite its 800-line complexity, it still behaves as an integral unit, rolling over from stable mode to stable mode as directed by its succession of input 60-tuples. Our concept employs the following design strategies: modular focusing of input information; modular decoupling under input changes; and modular redundancy of potential command (modules with the most information have the most authority). We have augmented these strategies to enable our model to condition, habituate, generalize, discriminate, predict, and generally follow a changing environment. We plan to further enrich the model to include en dogenous activity. Compared to cross-coupled perception arrays, our model is shown by symmetry arguments to compute different functions.