Locating and navigation mechanism based on place-cell and grid-cell models

Area of Science:

• Neuroscience
• Computational Neuroscience
• Cognitive Science

Background:

• Environmental cues are crucial for goal locating and navigation in rats.
• Previous navigation models primarily focused on place cells, neglecting the role of grid cells.
• Grid cells are known to contribute to spatial cognition and navigation.

Purpose of the Study:

• To develop a novel computational model for navigation integrating both place cells and grid cells.
• To simulate and analyze the mechanisms of learning and direction selection in spatial navigation.
• To investigate factors influencing locating accuracy and goal navigation performance.

Main Methods:

• Developed a continuous-time dynamic system incorporating place and grid cell interactions.
• Utilized simulation experiments to replicate physiological findings of place and grid cell firing fields.
• Analyzed the impact of parameters like learning rate, firing threshold, and cell number on navigation accuracy.

Main Results:

• Successfully reconstructed place fields of place cells and firing fields of grid cells in simulations.
• Identified key factors (learning rate, firing threshold, cell number) affecting navigation accuracy.
• Demonstrated convergence to a stable navigation path over multiple runs in a goal navigation task.

Conclusions:

• A integrated model of place and grid cells provides a more comprehensive understanding of navigation.
• The developed dynamic system effectively models learning and direction selection in spatial navigation.
• Simulation results align with physiological data, validating the model's predictive power for navigation behavior.