🔗 Long-Term Potentiation & LTD

Hebbian plasticity: synapse strength changes with correlated pre/post activity. BCM rule: dW/dt = φ(v_post)·v_pre where θ_M is a sliding modification threshold. See BCM curve and synaptic tagging.

NeuroscienceInteractive
BCM synaptic plasticity model · P pause · R reset · LTP/LTD buttons for stimulation bursts

How it Works

The BCM (Bienenstock-Cooper-Munro) rule models synaptic plasticity with a sliding threshold. The modification function φ(v_post) is negative (LTD) when postsynaptic activity is below θ_M and positive (LTP) when above. The threshold θ_M itself slides with the running average of postsynaptic activity, providing homeostatic stability.

The simulation shows the time course of synaptic weight W under different pre/post firing rate combinations. The BCM curve is plotted in real time, showing where LTP/LTD transitions occur.

dW/dt = φ(v_post, θ_M) · v_pre − γW
φ(v, θ) = v · (v − θ) [BCM modification function]
dθ_M/dt = (v_post² − θ_M) / τ_θ [sliding threshold]
v_post = W · v_pre + noise

Frequently Asked Questions

What is Long-Term Potentiation (LTP)?

LTP is a long-lasting increase in synaptic strength following high-frequency stimulation. It requires coincident pre- and postsynaptic activity (Hebbian rule), is dependent on NMDA receptor activation, and is widely considered a cellular mechanism for learning and memory formation.

What is Long-Term Depression (LTD)?

LTD is a long-lasting decrease in synaptic strength, typically induced by low-frequency stimulation or asynchronous pre/post activity. It is thought to balance LTP, preventing saturation of synaptic weights and allowing forgetting or memory refinement.

What is the Hebbian learning rule?

Hebb's rule: 'Neurons that fire together, wire together.' Formally, the synaptic weight change dW/dt ∝ v_pre · v_post. When pre- and postsynaptic neurons are simultaneously active, the synapse is strengthened. This is implemented biophysically through NMDA receptor coincidence detection.

What is the BCM rule?

The BCM rule extends Hebb: dW/dt = φ(v_post, θ_M) · v_pre, where φ changes sign at a sliding threshold θ_M. Below θ_M, activity induces LTD; above θ_M, it induces LTP. θ_M slides with mean postsynaptic activity, providing synaptic stability.

How do NMDA receptors act as coincidence detectors?

NMDA receptors require both glutamate binding (presynaptic) and postsynaptic depolarization to remove the Mg²⁺ block. This coincidence detection means only synapses active while the postsynaptic cell is depolarized become potentiated — implementing Hebbian plasticity at the molecular level.

What is spike-timing-dependent plasticity (STDP)?

STDP is a precise form of Hebbian plasticity where the direction and magnitude of synaptic change depends on the relative timing of pre- and postsynaptic spikes. If the presynaptic spike precedes postsynaptic firing by less than 20ms, LTP occurs; if it follows, LTD is induced.

What is synaptic tagging?

Synaptic tagging is a mechanism for late-LTP requiring protein synthesis. Brief stimulation sets a 'tag' at active synapses. When strong stimulation elsewhere triggers protein synthesis, those proteins are captured by tagged synapses, converting early-LTP (hours) into late-LTP (days), linking different memories.

How does LTP relate to memory?

LTP is the leading candidate for the synaptic basis of memory. Evidence includes: LTP induction uses the same protocols as learning, LTP occurs in memory areas (hippocampus), blocking NMDA receptors impairs both LTP and spatial learning, and spine enlargement during LTP mirrors changes seen in learning.

What is metaplasticity?

Metaplasticity is 'plasticity of plasticity' — the history of synaptic activity modifies the threshold for future LTP/LTD. In the BCM rule, θ_M slides upward with high average activity (making LTP harder to induce) and downward with low activity, preventing runaway potentiation.

What is the role of CaMKII in LTP?

CaMKII (Ca²⁺/calmodulin-dependent protein kinase II) is a key LTP effector. Ca²⁺ influx through NMDA receptors activates CaMKII, which phosphorylates AMPA receptors (increasing conductance), promotes AMPA receptor insertion into the synapse, and can remain active after Ca²⁺ returns to baseline (molecular memory).

About this simulation

This simulator runs the BCM (Bienenstock-Cooper-Munro) plasticity rule live: synaptic weight W evolves according to dW/dt = φ(v_post, θ_M)·v_pre − γW, where the modification function φ flips sign at a sliding threshold θ_M that itself tracks the running average of postsynaptic activity. Fire an ⚡ LTP Burst or ↓ LTD Stimulate and watch the weight trace, φ value, and sliding threshold respond in real time.

🔬 What it shows

The live evolution of synaptic weight W, the BCM modification function φ(v_post), and the sliding threshold θ_M under adjustable pre/post firing rates, plus discrete high-frequency (LTP) or low-frequency (LTD) stimulation bursts.

🎮 How to use

Set Pre-synaptic rate and Post-synaptic rate sliders, tune BCM threshold θ_M and Decay rate γ, then trigger ⚡ LTP Burst or ↓ LTD Stimulate to push postsynaptic activity above or below threshold. P pauses, R resets.

💡 Did you know?

The sliding threshold θ_M is what stops synapses from potentiating forever: because θ_M rises whenever average postsynaptic activity is high, a cell that just got potentiated actually becomes harder to potentiate further — a built-in homeostatic brake called metaplasticity.

Frequently asked questions

Why does clicking LTP Burst reliably increase the weight?

The burst temporarily forces both vpre and vpost to 80 Hz, well above the current θ_M, so φ(v_post) = vpost·(vpost−θ_M) becomes strongly positive, driving dW/dt up for the 3-second burst duration before rates return to their slider values.

Why does LTD Stimulate lower the weight instead?

Dropping both rates to 5 Hz pulls vpost below θ_M, making the term (vpost−θ_M) negative, so φ becomes negative and dW/dt turns downward — the same modification function produces depression rather than potentiation purely because activity now sits on the other side of the sliding threshold.

What happens to θ_M after repeated LTP bursts?

Because dθ_M/dt = (vpost² − θ_M)/τ_θ, sustained high postsynaptic activity from repeated bursts drags θ_M upward over ~10-second timescale, meaning the next LTP burst has to work harder to exceed the now-higher threshold — this is metaplasticity in action.

Why does the weight decay slowly even with no stimulation?

The −γW term in dW/dt is always active regardless of φ, representing a constant leak or forgetting rate; raising the Decay rate γ slider makes this leak faster, so the weight trace visibly drifts back toward baseline between bursts.

Why is φ defined as vpost·(vpost − θ_M) rather than just vpost − θ_M?

The extra vpost factor makes plasticity activity-dependent as well as threshold-dependent — a silent postsynaptic neuron (vpost≈0) produces almost no weight change regardless of θ_M, which matches the biological requirement that some postsynaptic activity is needed for any synaptic modification, LTP or LTD.