Action potentials are tiny electrical signals — just millivolts. To record them, scientists place a conducting electrode (metal or glass pipette filled with fluid) near or inside a neuron, amplify the signal, digitize it, and store it for analysis.

What encodes information in spikes?

Unlike graded potentials (variable size), all action potentials from the same neuron look identical: fixed amplitude, fixed duration (~2–4 ms). So the size of each spike carries no information.

Information is instead encoded in:
• Firing frequency — how many spikes per second (a neuron responding to a loud sound fires faster than one responding to a soft sound)
• Firing pattern — the temporal arrangement of spikes (some neurons burst briefly then go quiet; others fire at a steady regular rate; others at variable intervals)

Different neuron types have characteristic firing patterns, and these patterns change with the strength and type of stimulus.

Receptive fields:
Many neurons only respond to stimuli in a specific part of the world — their receptive field.

Example: some visual neurons fire strongly when light falls on the centre of their receptive field (on-centre) and weakly when light hits the surrounding ring (off-surround). Other cells do the opposite. This centre-surround organisation is how the retina detects edges and contrast, before the signal even reaches the cortex.

Plasticity:
Neural properties are not fixed. Learning, experience, drugs, and disease all reshape how neurons fire. In a diseased state, a neuron given the same stimulus may fire irregularly and reach a lower maximum frequency — revealing that the underlying electrical machinery has changed.

🔑 Key insight: because all spikes look the same, the only things that vary are timing and frequency. The entire richness of neural coding — everything you see, hear, feel, remember — is built from patterns of identical pulses.