Figure 6 gives an overview about the developed architecture for human-like machine perception which bases on insights about the working mechanisms of the human perceptual system. The central element of the model is the so-called “neuro-symbolic network”, which processes data coming from different sensor sources and additionally considers information coming from “higher-level” sources referred to as memory, knowledge, and focus of attention . Within the neuro-symbolic network, so called “neuro-symbolic information processing” takes place based on information exchange of “neuro-symbols”. The focus in this article will be on the description of the functioning of neuro-symbols and the neuro-symbolic network. Details about the other modules and functional aspects of the model can amongst others be found in.

https://www.edusoft.ro/brain/index.php/brain/article/view/419/472