Modelo de Desarrollo del Sistema Visual.

SOLETTA JH; LUCIANNA FA; PIZÁ AG; CANO LA; ALBARRACIN AL; FARFÁN FD

Jornada; XIV Jornadas de Ciencia y Tecnología de Facultades de Ingeniería del NOA; 2019

Some neurons of the primary visual cortex have special characteristics, such as selectivity to the orientation and direction of movement of visual stimuli. Although there are different models that try to explain the development of these characteristics, this is still a matter of debate. The use of computer models allows to integrate both the experimental data and the theoretical models proposed achieving a unified vision. In this work we have developed a computational model to study the possible learning mechanisms used by visual cortex neurons for the acquisition of selectivity in orientation and movement. The developed model is based on neural networks and contemplates the electrophysiological aspects of visual system neurons. The learning methods used were: Hebb, and Simple Genetic Algorithms. The models were evaluated by the construction of the tuning curve and compared with those obtained with cortical records in monkey bark. The results show that the learning method based on Genetic Algorithm allows obtaining tune curves similar to those obtained in cortical records.