AnimatLab Lateral Inhibition Network

This tutorial describes how to implement lateral inhibitory neural networks within the AnimatLab simulation environment (http://animatlab.com). AnimatLab is a neuromechanical simulation system that allows you to build a physically accurate, biomechanical model of the body an organism. Hill muscle models within that body can be controlled using biologically realistic neural networks to reproduce behaviors found in the real animals.

Lateral inhibition is an important circuit phenomenon that frequently occurs in situations where an array of sensory receptors maps the spatial distribution of a sensory stimulus.

Examples include retinal and sub-retinal circuits of the visual system, or the mechanosensory receptors detecting stimuli over the surface of the body. Lateral inhibition occurs when there is mutual inhibition between receptors responding to stimuli in adjacent parts of the receptive field.

Whatever the modality, the effect of lateral inhibition is to increase the contrast of edges by causing an enhanced differential in the response to the stimulus on either side of an edge.

Each layer of the lateral inhibition network serves to enhance the sensory signal so it is easier to detect. Neurons on either side of the signal are inhibited so their firing rate is driven below their normal value. lateral inhibition provides a nice contrast to let the nervous system determine where the signal originated, and to increase the resolution so it can tell the difference between two nearby signals.

This tutorial will show you how to create a very simple lateral inhibitory network to demonstrate these principles.

Видео AnimatLab Lateral Inhibition Network канала NeuroRoboticTech