Spatio-Temporal Pattern Representation from AI Inspired Brain Model in Spiking Neural Network

Abstract

Neuronal population activity in the brain is the combined response of information in the spatial domain and dynamics in the temporal domain. Modeling such Spatio-temporal mechanisms is a complex process because of the complexity of the brain and the limitations of the hardware. In this paper, we demonstrate how information processing principles adapted from the brain can be used to create a brain-inspired artificial intelligence (AI) model and represent Spatio-temporal patterns. The same is demonstrated by designing the tiny brain using spiking neural networks, where activated neuronal populations represent information in the spatial domain and transmitting signals represent dynamics in the temporal domain. Spatially located sensory neurons excited by input visual stimuli further activate motor neurons to trigger a motor response that causes behavior modification of the robotic agent. Initially, an isolated brain network is simulated to understand the excitation part from sensory to motor neurons while plotting waveform between membrane potential and time. The response of the network to stimulate robot body movements is also plotted to demonstrate representation. The simulation shows how the response of particular visual stimuli modifies behavior and helps us understand the body and brain synchronization. The perceived environment and resultant behavior response allow us to study body interaction with the environment.