How Lateral Interaction Develops in a Self-Organizing Feature Map

| A biologically motivated mechanism for self-organizing a neural network with modi able lateral connections is presented. The weight modi cation rules are purely activity-dependent, unsupervised and local. The lateral interaction weights are initially random but develop into a \Mexican hat" shape around each neuron. At the same time, the external inputweights self-organize to form a topological map of the input space. The algorithm demonstrates how self-organization can bootstrap itself using input information. Predictions of the algorithm agree very well with experimental observations on the development of lateral connections in cortical feature maps.