Neuronal synapse as a memristor: modeling pair- and triplet-based STDP rule

Area of Science:

• Neuroscience
• Materials Science
• Computational Neuroscience

Background:

• Synaptic plasticity, including long-term potentiation (LTP) and depression (LTD), is crucial for learning and memory.
• Existing models often simplify or omit the complex dynamics of synaptic weight modification.
• Spike-timing-dependent plasticity (STDP) is a fundamental mechanism governing synaptic efficacy based on the relative timing of pre- and postsynaptic spikes.

Purpose of the Study:

• To propose a new memristive synapse model that incorporates both short-term and long-term plasticity.
• To model higher-order synaptic behaviors using a memristor with adaptive thresholds, reflecting the Froemke suppression principle.
• To investigate the quantitative equivalence between the proposed memristive model and existing theoretical frameworks.

Main Methods:

• Development of a memristive model for neuronal synapses.
• Incorporation of spike-timing-dependent plasticity (STDP) protocols.
• Modeling of adaptive thresholds to simulate synaptic potentiation (LTP) and depression (LTD).
• Mathematical formulation using a set of ordinary differential equations.
• Quantitative comparison with Froemke's model for specific spike-timing scenarios ('pre-post-pre' and 'post-pre-post').

Main Results:

• The proposed memristive model successfully captures both long-term and short-term plasticity.
• Adaptive thresholds in the memristor model reproduce refractory periods in synaptic weight modification.
• The memristive model demonstrates quantitative equivalence to Froemke's model for 'pre-post-pre' and 'post-pre-post' spike patterns.
• A direct relationship between adaptive thresholds and short-term plasticity is established.

Conclusions:

• The novel memristive synapse model provides a unified framework for understanding synaptic plasticity.
• The model's ability to replicate known biological phenomena, such as refractory periods and equivalence to Froemke's principle, validates its approach.
• This work offers a promising direction for developing more biologically plausible artificial neural networks and neuromorphic computing systems.