On parsing the neural code in the prefrontal cortex of primates using principal dynamic modes

Area of Science:

• Neuroscience
• Computational Neuroscience
• Systems Neuroscience

Background:

• Analyzing functional connectivity in multi-input multi-output (MIMO) neuronal systems is complex.
• Traditional methods struggle to represent large-scale neural networks and their dynamic transformations.
• Understanding neuronal communication in the prefrontal cortex is crucial for cognitive tasks.

Purpose of the Study:

• To present a novel Principal Dynamic Modes (PDMs) based modeling methodology for MIMO neuronal systems.
• To analyze the dynamic transformations of neural activity from input (Layer 2) to output (Layer 5) in the primate prefrontal cortex.
• To investigate the relationship between neural activity patterns, brain rhythms, and behavioral performance.

Main Methods:

• Applied Principal Dynamic Modes (PDMs) to nonlinear modeling of multi-unit recordings.
• Analyzed spike train activity from Layer 2 to Layer 5 in the prefrontal cortex of non-human primates.
• Correlated PDM activation with behavioral outcomes during a Delayed-Match-to-Sample task.

Main Results:

• PDM-based models successfully reduced the complexity of large-scale neural MIMO systems.
• Identified "input-output channels of communication" linked to specific neural rhythm frequency bands.
• Found differential activation of frequency-specific PDMs associated with correct versus incorrect task performance.

Conclusions:

• Principal Dynamic Modes (PDMs) provide a powerful tool for analyzing functional connectivity and dynamic transformations in complex neuronal networks.
• PDM analysis reveals frequency-specific communication channels within the prefrontal cortex.
• Neural activity patterns, as captured by PDMs, are modulated by cognitive task demands and behavioral outcomes.