When biological neurones pass potential to one another, they do not do so in lock step and using continuous values, as is usually the case in {[artificial neutal networks}}, but as short spikes of ionic potential. These are more likely to occur when a cell is highly activated, leading to more frequent spikes, and the continuous value approxmates this, but the spiking can lead to different behaviours. This more detail bilogical model is the bases of spiking neural networks.
Used on page 123
Links:
lcnwww.epfl.ch: Spiking Neuron Models: Single Neurons, Populations, Plasticity
- neuronaldynamics.epfl.ch: Neuronal Dynamics: From single neurons to networks and models of cognition