firingrate

Firing rate is a basic descriptor of neuronal activity. It refers to the average number of action potentials, or spikes, produced by a neuron per unit time, and is most commonly expressed in spikes per second or hertz. In experimental data, firing rate is used to summarize how a neuron responds to stimuli or to ongoing activity.

Firing rate can be estimated from spike trains by counting spikes within a time window and dividing by the window length. Researchers may also derive time-varying rates using histograms, kernel smoothing, or model-based approaches. The instantaneous firing rate is a continuous estimate of spike probability over time.

Firing rate is subject to variability. Even with a constant input, successive trials show different spike counts due to the stochastic nature of spiking, network fluctuations, and adaptation. The Poisson model is a simple reference in which spikes are drawn with a fixed rate, but real neurons exhibit refractoriness, bursting, and non-Poisson statistics.

In coding theories, firing rate is a primary carrier of information in rate coding, where the stimulus or decision variable is represented by average firing rate over a relevant window. Other coding schemes emphasize precise spike timing or synchrony, which convey information beyond mean rate.

Neuronal models link firing rate to input and intrinsic properties. Biophysical models such as Hodgkin-Huxley or mushroom body-type models describe spike generation, while rate-based models describe the evolution of firing rate with differential equations and synaptic inputs.