New trends in computational modeling: a Neuroid-based retina model.

It is thought that using detailed neuron-models could lead to a better understanding of how the nervous system works. However, neural networks preserve their collective computational properties, regardless of the level of description used for modeling the main building block. In this paper, we built a Neuroid-based retina model. As a result of the implementation, the Neuroid was able to reproduce the essential features of the photoreceptor response to light. Likewise, the retina model responded to specific visual effects such as simultaneous contrast, Mach bands and Hermann grid. All of these suggest that the Neuroid comprises enough functional characteristics, such that we could focus not only on the most relevant computational aspects of nerve cells, but also in the collective capabilities of large-scale neural networks.