Spike Timing Neural Model of Motion Perception and Decision Making

Area of Science:

• Computational Neuroscience
• Systems Neuroscience
• Artificial Intelligence

Background:

• Human decision-making in visual navigation involves complex neural processing.
• Simulating these processes requires sophisticated computational models.

Purpose of the Study:

• To develop and validate a hierarchical spike timing neural network model in NEST.
• To reproduce human decision-making in simplified simulated visual navigation tasks.
• To investigate the neural mechanisms underlying saccade generation based on optic flow.

Main Methods:

• Hierarchical spike timing neural network model implemented in NEST simulator.
• Multi-layered architecture including retinal, thalamic, and cortical areas (V1, MT, MST, LIP).
• Simulated decision-making using mutually inhibitory sub-regions in the LIP layer.
• Stimulus-response testing at each model stage and overall performance evaluation with optic flow patterns.

Main Results:

• The model successfully processed visual information through hierarchical layers.
• Specific stimulus selectivity was validated for V1, MT, and MST areas.
• The LIP layer's decision-making component accurately correlated firing rates with simulated saccade choices (left/right).
• The model demonstrated performance in reproducing self-motion perception from optic flow.

Conclusions:

• The developed hierarchical neural network model effectively reproduces key aspects of human visual navigation and decision-making.
• The model provides a valuable tool for understanding the neural basis of spatial cognition and motor responses.
• Further research can extend this model to more complex navigation scenarios and decision-making processes.