Extracellular Spike Recording: Volume-Conductor 1/r Decay and Probe Source Localization
A neuron fires beside a 16-site silicon probe in tissue modelled as a homogeneous conductor. The spike amplitude falls as 1/r from the current sink, so the pattern of amplitudes across the contacts tells you where the neuron is. Change distance, depth, current, conductivity, noise and the return-current line, then see the detection radius, a Monte Carlo cloud of position estimates and the Cramer-Rao bound, with numerical checks built in.
A neuron fires beside a 16-site silicon probe in tissue modelled as a homogeneous conductor. The spike amplitude falls as 1/r from the current sink, so the pattern of amplitudes across the contacts tells you where the neuron is. Change distance, depth, current, conductivity, noise and the return-current line, then see the detection radius, a Monte Carlo cloud of position estimates and the Cramer-Rao bound, with numerical checks built in.
A neuron fires beside a 16-site silicon probe in tissue modelled as a homogeneous conductor. The spike amplitude falls as 1/r from the current sink, so the pattern of amplitudes across the contacts tells you where the neuron is. Change distance, depth, current, conductivity, noise and the return-current line, then see the detection radius, a Monte Carlo cloud of position estimates and the Cramer-Rao bound, with numerical checks built in.
3D · Three.js / WebGL renderer · 60 FPS target · runs fully client-side, no install