A stochastic method to predict the consequence of arbitrary forms of spike-timing-dependent plasticity

Area of Science:

• Neuroscience
• Computational Neuroscience
• Biophysics

Background:

• Synaptic plasticity, specifically spike-timing-dependent plasticity (STDP), is crucial for neural computation.
• STDP learning rules vary across neural preparations, impacting synaptic strength.
• Existing models often simplify STDP learning windows, limiting predictive power.

Purpose of the Study:

• To develop a rigorous mathematical framework for calculating synaptic conductance distributions under various STDP learning window functions.
• To analyze the impact of different STDP rules on synaptic competition and distribution.
• To identify STDP window functions that optimize synaptic competition.

Main Methods:

• Utilized the Fokker-Planck equation to model conductance distributions.
• Employed the Ornstein-Uhlenbeck process to simulate membrane potential fluctuations.
• Validated the method against known STDP learning outcomes.

Main Results:

• Demonstrated that STDP with a single-exponential window can lead to bimodal conductance distributions.
• Showed that minor modifications to the window function can yield unimodal distributions.
• Applied the method to hippocampal and electric fish STDP, revealing optimized competition and all-or-none synapse scenarios.

Conclusions:

• The developed mathematical method accurately predicts synaptic conductance distributions for diverse STDP rules.
• Specific STDP mechanisms, like those in the hippocampus, promote efficient synaptic competition.
• Other STDP types, like in electric fish, can lead to extreme synaptic configurations.