Dymods: A framework for modularizing dynamical neuronal structures

We present an informal framework for modularizing dynamical systems so that they can be recombined to form larger dynamical structures. We use this fr amework to give a cellular account of how network dynamics might be respons ible for the maintenance of reversal locomotion in the nematode C. elegans. Three nonlinear dynamical modules were constructed using physiologically r ealistic cellular models and they were assembled to form the complete circu it. Finally, we present a real-time, distributed implementation framework t hat allows dynamical modules to be implemented independently an multiple co mputer systems and easily interconnected. (C) 1999 Elsevier Science B.V. Al l rights reserved.