Project replicating the spike-timing-dependent plasticity (STDP) framework proposed by Chindemi et al. (2022) inside the NESTML simulation environment.
Synaptic plasticity is the ability of synapses to strengthen or weaken over time as a function of spike timing. The calcium-control hypothesis suggests that postsynaptic calcium dynamics determine whether long-term potentiation (LTP) or long-term depression (LTD) occurs.
In this project, we modeled neurons and synapses in NESTML/PyNestML to reproduce the calcium-based STDP mechanism:
-
Neuron model:
Based on the Hill–Tononi point-neuron formalism, simplified for computational efficiency while retaining key synaptic currents (AMPA, NMDA). -
Synapse model:
Custom implementation of calcium-dependent plasticity. Calcium influx is modeled through:- NMDA receptor current (including magnesium block),
- Voltage-dependent calcium channels (VDCC, after Chindemi et al.).
-
Plasticity mechanism:
- Calcium traces integrated through a leaky accumulator (c⋆, per Chindemi et al.).
- Synaptic efficacy (ρ) updated once potentiation or depression thresholds are crossed.
- ρ modulates AMPA conductance and presynaptic release probability.
- Vesicle release simplified from the original stochastic model to a mean weight with Gaussian noise.
-
Pre-before-post spiking:
Temporal alignment of NMDA receptor activation with postsynaptic depolarization produces supralinear calcium transients, leading to LTP. -
Post-before-pre spiking:
Misaligned events yield smaller calcium transients, sufficient only for LTD. -
Frequency dependence:
Longer NMDA-mediated calcium events dominate shorter VDCC ones, captured by the leaky calcium integrator.
mhill_tononi_neuron.nestml— NESTML implementation of the neuron model.stdp_ca_synapse.nestml— NESTML implementation of the calcium-based STDP synapse.notebook.ipynb— Jupyter notebook with the simulations.supplementary_material.pdf— Comments on findings.