NEURAL NET MODEL OF THE GILL-WITHDRAWAL REFLEX IN APLYSIA

The overwhelming complexity of CNS of phylogenetically higher species has contributed to the interest for invertebrate's CNSs neurophysiology demonstrates. Moreover, the existence of many behavioral-bound neuronal features common along phylogenesis appears to-days more than likely. Many cellular mechanisms held responsible for habituation (a simple form of learning) and sensitization, have been early elucidated in the neuronic implementation of the gill-withdrawal reflex in Aplysia. This paper describes a model of such a neural system: for most of the synapselike elements a kind of hebbian rule has been adopted. Together with the illustration of these concepts a computer simulation showing the model's capacity to habituate and to react in a biologically plausible way to various stimuli is described. As long as habituation and sensitization are the basic requirements for learning by conditioning, the model appears suitable as a modular unit in a larger device apt for these tasks. It may also be found useful in simulations aiming at elucidating "relevance-circuit" - type mechanisms.