The spatio-temporal encoding of the information by the neurons of the frog's 'visual brain' and model representations of the multiparameters analysis of the features of the visual stimuli by the neural networks

Information encoding and processing in different parts of the frog visual cortex have been investigated. Experiments were performed on frogs (Rana ridibunda) immobilized with tubocurarine (after prior local injections of novocaine). Impulse activity was recorded extracellularly from retinal ganglion cells and from neurons of mesencephalon (tectum and tegmentum), of diencephalon (nucleus posterior thalmi and Bellonci nucleus) and of telencephalon (primordial hippocampus). The following types of visual stimuli were used: diffuse illumination, stationary and moving light, and dark spots and bars. It has been revealed that neurons in the visual cortex transfer information using various possible codes: positional coding by indicator lines, various modifications of temporal codes, and ensemble coding. The experimental results have made it possible to propose models of central neuronal networks. These models may be constructed on the basis of both narrowband and wideband retinal inputs, or only on the basis of wideband inputs.<
>