Caltech Library System

About | Browse | Search | Caltech Student Instructions

Type of Document	Dissertation
Author	Gold, Carl
URN	etd-05312007-210112
Persistent URL	http://resolver.caltech.edu/CaltechETD:etd-05312007-210112
Title	Biophysics of extracellular action potentials
Degree	PhD
Option	Computation and Neural Systems
Advisory Committee	Advisor Name Title Richard Andersen Committee Chair Christof Koch Committee Member Gilles Laurent Committee Member Gyorgy Buzsaki Committee Member John Allman Committee Member
Keywords	neuron simulation extracellular recording model CA1 V1
Date of Defense	2007-05-18
Availability	unrestricted
Abstract The goal of this thesis is to analyze the generation of single unit extracellular action potentials (EAPs), and to explore pertinent issues in the interpretation of EAP recordings. I use the line source approximation to model the EAP produced by individual neurons. I compare simultaneous intracellular and extracellular recordings of CA1 pyramidal neurons in vivo with simulations using the same cells'reconstructions. The model accurately reproduces both the waveform and the amplitude of the EAPs. The composition of ionic currents is reflected in the features of each cell's EAP, while dendritic morphology has little impact. I compared constraining a compartmental model to fit the EAP with matching the intracellular action potential (IAP). I find that the IAP method underconstrains the parameters. The distinguishing characteristics of the EAP constrain the parameters and are fairly invariant to electrode position and cellular morphology. I conclude that matching EAP recordings are an excellent means of constraining compartmental models. I recorded spikes from cat primary visual cortex (V1) and recreated them in the model. I calculated the distance at which an electrode could record the EAPs given the prevalent background noise. My analysis suggests that in the superficial cortical layers 50%-80% of the neurons were active, while in deeper layers only 10%-20% were active. I analyzed the bias towards recording the large neurons in the deep layers. If the detection and clustering algorithm is sensitive enough to include low-amplitude spikes then bias is moderate. If only high amplitude units (> 0.2 mV) are picked up, then recording will be significantly biased towards the deep layers. The majority of spikes in cortex had a negative peak with a mean of -0.11 mV, but a minority of units (<10%) had a large positive peak of up to 1.5 mV. Simulations demonstrate that a pyramidal neuron may generate a negative spike with amplitude greater than 1 mV, but a positive spike of at most 0.5 mV. I conclude that high-amplitude positive spikes cannot result from a single neuron EAP. I suggest that they may result from synchronized action potentials in groups of L5 pyramidal neurons.
Files	Filename       Size       Approximate Download Time (Hours:Minutes:Seconds)     28.8 Modem   56K Modem   ISDN (64 Kb)   ISDN (128 Kb)   Higher-speed Access    cgold_thesis_2.pdf 17.72 Mb 01:22:02 00:42:11 00:36:54 00:18:27 00:01:34