Comparison and regulation of neuronal synchronization for various STDP rules

Area of Science:

• Computational Neuroscience
• Neuroscience
• Artificial Intelligence

Background:

• Spike-timing-dependent plasticity (STDP) is a key mechanism for synaptic plasticity in neural systems.
• Understanding STDP's role in neural network dynamics, like synchronization, is crucial for brain function and AI.
• Previous research has explored STDP's impact on neural activity, but systematic analysis across various rules is limited.

Purpose of the Study:

• To systematically investigate the effects of different experimentally supported STDP learning rules on the frequency synchronization of two unidirectionally coupled neurons.
• To compare the synchronization capabilities of STDP rules against a constant connection baseline.
• To identify optimal STDP learning parameters for maximizing synchronization windows.

Main Methods:

• Simulated two unidirectionally coupled neurons.
• Implemented and tested various experimentally supported Spike-Timing-Dependent Plasticity (STDP) learning rules.
• Analyzed the synchronization window (range of frequencies for synchronized firing) for each STDP rule.
• Varied STDP learning parameters to find optimal configurations.

Main Results:

• No tested STDP learning rule enhanced the synchronization window compared to a constant connection under identical model conditions.
• The synchronization window's width was found to be highly dependent on the specific shape and parameters of the STDP update rules.
• Optimal STDP parameters were identified that resulted in the widest synchronization windows achievable with STDP.

Conclusions:

• The effectiveness of STDP in promoting neural synchronization is not universal and is critically dependent on the specific learning rule and its parameters.
• Constant connections provide a robust baseline for synchronization, suggesting that STDP's role may lie in other aspects of network adaptation.
• Further analysis of synchronization mechanisms underlying different STDP rules is warranted to fully understand their impact on neural network dynamics.