Engineering spatiotemporal patterns: information encoding, processing, and controllability in oscillator ensembles.

Area of Science:

• Neuroscience
• Control Theory
• Computational Biology

Background:

• Precise manipulation of neuronal population activity is crucial for understanding brain functions, sleep, and neurological disorders.
• Challenges in neural control include the scale of neuronal networks and the lack of real-time state measurements, hindering performance.

Purpose of the Study:

• To formulate neural population control as a tracking problem.
• To propose a principled, feedback-free control technique for generating desired spatiotemporal patterns in neuronal ensembles.
• To explore information encoding and processing via controllability properties in neuron ensembles.

Main Methods:

• Formulated the control of dynamic structures in neuron ensembles as a tracking problem.
• Developed an optimal stimulus design technique that does not require feedback information.
• Validated the technique on mathematical models (Kuramoto, Hodgkin-Huxley) and real-time experiments (Wein bridge oscillators).

Main Results:

• Demonstrated the ability to design optimal stimuli for producing desired spatiotemporal patterns in neuronal networks.
• Showcased the technique's effectiveness in creating complex spatiotemporal spiking patterns.
• Revealed insights into information encoding and processing through the lens of controllability in neuron ensembles.

Conclusions:

• The proposed feedback-free control technique effectively generates desired spatiotemporal patterns in neuronal populations.
• This approach offers a powerful tool for neuroscience research and potential therapeutic interventions for neurological conditions.
• Controllability analysis provides a novel perspective on information processing within neural ensembles.