Coincidence Detection

Coincidence detection is the neural computation in which a neuron fires only when multiple synaptic inputs arrive within a narrow temporal window — typically a millisecond or less. It is the brain's mechanism for detecting synchrony and converting temporal correlation into spatial maps. The canonical example is sound localization in the medial superior olive, where microsecond differences in acoustic arrival time between the two ears are converted into a topographic representation of sound source direction.

Coincidence detection depends on the biophysical properties of the postsynaptic neuron: a steep membrane potential rise time and a rapid decay of synaptic currents. These properties make the neuron a coincidence detector rather than an integrator. The distinction between coincidence detection and temporal integration is central to neural coding: integrators average; detectors resolve. And resolution is what makes temporal coding possible. The same principle operates in electronic phase-locked loops, where a phase comparator produces output only when reference and feedback signals are temporally aligned.