Temporal coding

The temporal coding is a kind of neural coding in which a neuron encodes information through the accurate timing of action potentials, or spikes, on a time scale of millisecond. There is no general definition of temporal coding; however efforts have been made to distinguish the accurate timing of spikes in a single neuron that encodes information about a stimulus from synchronized firing of neurons within a localized region. The later is sometimes termed to as correlation coding.

The temporal encoding might be implemented by single neurons in the cercal sensory system of the house cricket Acheta domesticus. Such neurons display a greater-than-expected coding capacity, due in section to an increased precision in short pattern of the action potentials. We generated linear and non-linear models for decoding the activity of such neurons. The stimuli related with short-interval patterns of spikes (ISIs of 8 ms or less) could be predicted better by second-order models as compared to the linear models. Lastly, we characterized the difference among these linear and second-order models in a low-dimensional subspace, and displayed that alteration of the linear models along only a few dimensions improved their predictive power to parity with the second order models. Altogether these outcomes show that the single neurons are able of using temporal patterns of spikes as primary symbols in their neural code, and that they communicate particular stimulus distributions to consequent neural structures.