Related Experiment Videos
Uncovering the synchronization dynamics from correlated neuronal activity quantifies assembly formation
J Deppisch1, K Pawelzik, T Geisel
1Institut für Theoretische Physik, Universität Frankfurt, Frankfurt/Main, Germany.
Biological Cybernetics
|January 1, 1994
Summary
Detecting neuronal network synchronization is challenging due to neuron stochasticity. This study introduces a novel hidden state model to identify synchronized neuronal assemblies and their dynamics from spike train data.
Area of Science:
- Neuroscience
- Computational Neuroscience
- Systems Neuroscience
Background:
- Synchronous network excitation is crucial for neuronal information processing.
- Detecting synchronized neuronal states in electrode recordings is difficult due to the stochastic nature of neurons.
Purpose of the Study:
- To present a framework and model for identifying network states and their dynamics.
- To operationalize the concept of neuronal groups forming assemblies via synchronization.
- To quantify neuronal activity based on internal network phase.
Main Methods:
- Developed a hidden state model, formally equivalent to a hidden Markov model.
- Utilized the Baum-Welch algorithm for accurate parameter determination from experimental time series.
- Applied the method to cat visual cortex recordings with oscillations and synchronizations.
Main Results:
- The hidden state model successfully identified synchronization, oscillation, switching, background activity, and correlations.
- Parameters uncovered characteristic system properties.
- Quantified assembly formation and enabled precise localization of system states.
Conclusions:
- The framework effectively identifies and characterizes synchronized neuronal network states and their dynamics.
- The model provides a powerful tool for analyzing complex neural activity patterns.
- Applicable to multielectrode recordings for understanding neuronal assembly formation.