tuning curve

A tuning curve is a simple graph that shows how strongly a single neuron reacts as you slowly change one feature of what it is sensing. Imagine turning a radio dial: most stations are static, but at one spot the music comes in loud and clear. A neuron often behaves the same way. As you sweep a stimulus across its possible values — the angle of a line you show the eye, the pitch of a tone, the direction an arm points — the neuron stays fairly quiet for most settings and then fires fast at one favorite value. Plot the stimulus along the bottom and the neuron's firing rate up the side, and you usually get a hill-shaped curve that rises to a peak and falls off on either side.

That peak is the neuron's preferred stimulus, and the width of the hill tells you how picky it is: a narrow, sharp curve means the cell cares about a very specific value, while a broad, gentle curve means it responds to a wide range. Scientists love tuning curves because they reveal what a neuron is for and how the brain encodes information. Since no single fussy cell can pin down a value on its own, the brain reads many neurons together — each with its peak in a slightly different place — and combines their votes. The strongest and the in-between responses across the whole population point to the answer, a strategy known as population coding.