Heteroclinic contours in oscillatory ensembles

Area of Science:

• Computational Neuroscience
• Dynamical Systems Theory
• Neural Network Modeling

Background:

• Sequential activity in neural ensembles is crucial for cognitive functions like memory and decision-making.
• The winnerless competition (WLC) principle offers a theoretical framework for understanding sequential dynamics.
• Existing models often simplify the complex dynamics of neural networks.

Purpose of the Study:

• To investigate the emergence of sequential activity in inhibitory coupled neuronlike oscillators.
• To identify the critical bifurcations that lead to the formation of stable heteroclinic sequences and channels.
• To link the WLC principle to observable dynamical behaviors in neural network models.

Main Methods:

• Simulating ensembles of neuronlike oscillators with inhibitory coupling.
• Analyzing phase space dynamics to identify stable heteroclinic sequences and channels.
• Investigating bifurcations that trigger sequential activity and heteroclinic structures.

Main Results:

• Demonstrated the occurrence of stable heteroclinic sequences and channels in oscillatory neural models.
• Identified specific bifurcations responsible for the onset of sequential and metastable dynamics.
• Confirmed the relevance of the WLC principle in generating complex neural activity patterns.

Conclusions:

• The WLC principle provides a robust mechanism for generating sequential activity in neural networks.
• Bifurcation analysis is key to understanding transitions to sequential dynamics.
• Oscillatory network models can exhibit complex metastable dynamics relevant to neural computation.