Self-organization, Learning and Language

Several issues related to Learning process are discussed from the viewpoint of self-organization in this paper. Though Hebbian rule and BCM rule are widely used in learning process, there may be a more general mechanism behind the rules and J structure with specific attractors may be such kind of mechanism. Chinese character learning is a good example of learning process. A neural network based on Hebbian rule is developed to learn Chinese grapheme. The results show that the Chinese character can be learned with appropriate parameters and integration method. Two properties in Chinese learning at behavioral level are discussed. However, at the level of neurons and neuron groups, a small network with dynamic synapse was put forward by Berger and the complex cognitive activities such as sound recognition can be achieved by such simple neural network. The supposition that there should be basic mechanism to govern cognitive activities is partly validated. Furthermore, information is involved in any learning process. How information changes is the core problem of learning process and still open. Some discussion on this problem is also given in this paper.