Question 1

What is synaptic plasticity?

Accepted Answer

Synaptic plasticity is the ability of the connection (synapse) between two neurons to change in strength over time. When the connection grows stronger a presynaptic spike has a bigger effect on the postsynaptic neuron; when it weakens the effect shrinks. This change in synaptic weight is widely believed to be the cellular basis of learning and memory.

Question 2

What does STDP stand for?

Accepted Answer

STDP stands for spike-timing-dependent plasticity. It is a learning rule in which the precise relative timing of pre- and postsynaptic spikes, measured in milliseconds, determines whether a synapse is strengthened (potentiated) or weakened (depressed).

Question 3

How does spike timing change the synapse?

Accepted Answer

If the presynaptic neuron fires shortly before the postsynaptic neuron (dt = t_post − t_pre > 0) the synapse is potentiated, so the weight increases. If the presynaptic neuron fires shortly after the postsynaptic one (dt < 0) the synapse is depressed and the weight decreases. The closer the spikes are in time, the larger the change.

Question 4

What is the STDP equation?

Accepted Answer

For a spike pair with timing difference dt = t_post − t_pre: dw = A+ · exp(−dt/tau+) when dt > 0 (potentiation), and dw = −A− · exp(dt/tau−) when dt < 0 (depression). A+ and A− set the maximum step sizes and tau+ and tau− (typically around 20 ms) set how quickly the effect decays with timing.

Question 5

What is a leaky integrate-and-fire (LIF) neuron?

Accepted Answer

A leaky integrate-and-fire neuron is a simple model where the membrane voltage integrates incoming current but also leaks back toward a resting value. When the voltage crosses a threshold the neuron emits a spike and resets. It is the standard minimal model used to study spike timing and plasticity.

Question 6

What is Hebbian learning?

Accepted Answer

Hebbian learning is summarised as 'neurons that fire together wire together'. STDP is a precise, timing-based form of Hebbian learning: it rewards a presynaptic neuron that reliably helps cause the postsynaptic neuron to fire, and punishes one that fires too late to have contributed.

Question 7

Why is the order of firing important?

Accepted Answer

Causality matters. A presynaptic spike just before a postsynaptic spike could have helped cause it, so strengthening that link is useful for prediction and learning. A presynaptic spike just after cannot have caused the output, so weakening it removes a misleading connection. STDP captures this causal asymmetry.

Question 8

What do tau+ and tau- control?

Accepted Answer

tau+ and tau− are the time constants of the potentiation and depression sides of the STDP window. Larger time constants mean the synapse responds to spike pairs separated by longer intervals; smaller values make plasticity sensitive only to very tightly synchronised spikes.

Question 9

Does the synapse keep growing forever?

Accepted Answer

No. In this simulation, and in real neurons, the weight is bounded between a minimum and a maximum. Without bounds STDP could run away, so biological synapses use saturation, weight-dependent rules and homeostatic mechanisms to keep activity stable.

Question 10

How is STDP used in artificial neural networks?

Accepted Answer

STDP is the main local, unsupervised learning rule for spiking neural networks and neuromorphic hardware. Because each synapse updates using only the timing of its own pre- and postsynaptic spikes, it scales efficiently and is attractive for low-power, brain-inspired computing chips.

🧠 Synaptic Plasticity — How Neurons Learn (STDP)

What did you find?