Computational modeling and analysis of hippocampal-prefrontal information coding during a spatial decision-making...

Area of Science:

• Computational neuroscience
• Behavioral neuroscience
• Systems neuroscience

Background:

• Understanding the neural basis of decision-making is crucial for cognitive neuroscience.
• Spatial navigation and working memory are key components of complex decision-making tasks.
• Previous research suggests distinct roles for the medial prefrontal cortex (mPFC) and dorsal hippocampus (dCA1) in these processes.

Purpose of the Study:

• To develop and validate a computational model of rat behavior in a maze-based decision-making task.
• To investigate the neural mechanisms underlying spatial and mnemonic information processing during decision-making.
• To predict and analyze neuronal activity patterns in the mPFC and dCA1.

Main Methods:

• Development of a computational model integrating sensory input and working memory for decision-making.
• Behavioral analysis of rat performance in a maze task.
• Neurophysiological recordings of mPFC and dCA1 neuronal activity.
• Utilizing a novel software toolbox (Maze Query Language, MQL) for data analysis.

Main Results:

• The model successfully reproduced rat behavioral data.
• dCA1 neuronal firing rates discriminated the context (previous turn direction).
• A subset of mPFC neurons showed selectivity for current turn direction or context, with some encoding both.
• mPFC neurons exhibited ramping activity approaching decision points, with reduced selectivity during error trials.

Conclusions:

• The computational model provides a framework for understanding neural computations in decision-making.
• Neuronal activity in dCA1 and mPFC plays distinct but complementary roles in spatial navigation and choice behavior.
• Findings align with primate studies, suggesting conserved neural mechanisms for evidence integration in decision-making across species.