Divide et impera: optimizing compartmental models of neurons step by step

Compartmental models of neurons, introduced by Wilfrid Rall in 1964, have become important research tools for both theoretical and experimental neuroscientists to describe electrical (and sometimes also chemical) signalling in neurons. Usually built based on experimental data, they in turn help to interpret experiments, provide a quantitative description of neuronal function in contexts which are not yet directly accessible to experiment, and guide the development of theories of information processing in neurons and neural circuits. Detailed compartmental models, which incorporate anatomical reconstructions of the morphology of a neuron, biophysical descriptions of the kinetics and voltage dependence of its membrane conductances, as well as the membrane capacitance and intracellular resistivity, can predict the evolution in time and space of the membrane potential along the neuronal dendrites, soma and axon in response to arbitrary spatiotemporal patterns of synaptic input or current injection via intracellular electrodes. They can also form the basis for the construction of reduced, or simplified, neuronal models (reviewed by Herz et al. 2006).