Neural network representing apparatus having self-organizing function

An integrated circuit device being modelled on nerve cells, comprising:

• a plurality of neuron-expressing units simulating functions of neurons of nerve cell bodies;
• a plurality of axon-representing signal lines each receiving a signal indicating a state of an associated neuron-representing unit;
• a plurality of dendrite-representing signal lines corresponding to respective ones of said plurality of neuron-expressing units for transferring signals to corresponding neuron-expressing units;
• a plurality of synapse-expressing units on respective crosspoints between said plurality of axon-representing signal lines and said plurality of dendrite-representing signal lines coupling related axon-representing signal lines with related dendrite-representing signal lines with synapse load values indicating connecting strength between an axon-representing signal line and a dendrite-representing signal line, said integrated circuit device operable in a self-organizing manner of correcting a synapse load value in response to externally supplied educator information in accordance with a prescribed learning rule, said learning rule including a synapse load correction coefficient defining an amount of the synapse load value corrected by a single learning operation, said synapse load correction coefficient being a positive constant commonly determined for all said synapse-expressing units;
• and
• means for changing said connecting strength indicated by each respective synapse load value to a quantity which is a function of a distance between two neuron-expressing units coupled by each respective synapse-expressing unit.