A Bioinspired Model of Decision Making Considering Spatial Attention for Goal-Driven Behaviour

Abstract

Cognitive architectures (CA) are currently used to approach computer systems’ behavior to human behavior and intelligence. Fundamental human capability is planning and decision-making. In that regard, numerous AI systems successfully exhibit human-like behavior but are limited to either achieving specific objectives or too heavily constrained environments, which makes them unsuitable in the presence of unforeseen situations where autonomy is required. In this work, we present a bioinspired computational model to undertake the autonomous navigation problem as a result of the interaction between planning and decision-making, spatial attention and the motor system. The proposed model is embedded in a greater cognitive architecture. In the case study developed, it is proposed and tested that the process of planning and decision-making plays an important role to carry out spatial navigation. In it, the agent must move through an unexplored maze from an initial point to a final point, which it accomplished successfully. The gathered results prompt us to continue working on the model that considers attentional information to guide the agent’s behavior, which is strongly supported by the concise selection of neuroscientific evidence related to the cognitive functions we provided.