Top Highlights

• We cover the morphology loop with a large number of identical but translated functions which we call the CSD basis functions denoted by b˜j(𝐱) 𝑏 ~ 𝑗 ⁢ ( 𝐱 ) . There is a large flexibility here, but in practice we use Gaussians, so b˜j(𝐱)∝exp(−(𝐱−𝐱j)2/2R2) 𝑏 ~ 𝑗 ⁢ ( 𝐱 ) ∝ exp ⁡ ( - ( 𝐱 - 𝐱 𝑗 ) 2 / 2 ⁢ 𝑅 2 ) . The number of basis functi...
• Computing spike-triggered average of the potential, which we do in our proof-of-concept experiment, or driving the neuron with sinusoidal current and averaging the extracellular potential over periods of the driving current, are ways in which this could be achieved
• we study the general quality of reconstruction of fine detail by considering CSD distributions in the form of standing waves of increasing spatial frequency which form the Fourier basis of any possible CSD profil
• For now, we shall ignore the challenge of separating the part of the signal contributed by the studied neuron from background extracellular signals generated by nearby cells; we shall return to this issue in the Discussion.
• CSD basis functions

Domain