Neuroscientists use computer simulations of neural systems in their efforts to understand processes that underlie neural function. As experimental data in- crease, it becomes clear that detailed physiological data alone are not enough to infer how neural circuits work. Experimentalists appear to be recogniz- ing the need for a quantitative approach to the exploration of the functional consequences of particular neural features, which is provided by modelling. The number of computer simulation programs is designed as a tool for de- velopment and simulation of realistic models of single neurons and neural networks. The present available packages for modelling of biological neural networks are often dedicated Unix-based simulation packages, which require rather large computational power from workstations, typically Unix systems. The widely distributed packages, as Genesis [8] and Neuron [4], have their own interpreted scripting language, in which users define components and run- ning parameters for their simulations. In the hands of experienced users with access to a compatible computer system, these modelling packages are powerful research tools. However, they do suffer several drawbacks for non- expert users: they don't provide a Graphical User Interface (GUI) or have a very simple one, and as a result of it they can't visually represent the simula- tion process. Also, the formal structure of the language is difficult and time consuming to learn; at least initial knowledge and skills about Unix system are necessary for users.
