Biophysical Source Modeling of Some Exogenous and Endogenous Components of the Human Event-Related Potential

Bioinformation Systems;Engineering Science

Eriksen, K. Jeffrey ; Fender, Derek H.

University:California Institute of Technology

Department:Engineering and Applied Science

关键词: Bioinformation Systems;    Engineering Science;   

Others  :  https://thesis.library.caltech.edu/4439/1/Eriksen_kj_1984.pdf

美国|英语

来源: Caltech THESIS

【 摘 要 】

Methods of dipole localization were applied to human scalp-recorded electrical activity associated with a simple auditory cognitive discrimination task.

Human neuroanatomy and neurophysiology were reviewed from a biophysical standpoint in order to describe the probable neurogenesis of electrical activity in the brain and on the surface of the head. Topographic electroencephalography (EEG) analysis and source localization methods were historically reviewed in detail, followed by a brief review of the history of non-invasive evoked potential (EP) and magnetic field measurements of human central nervous system activity.

Four well known simple cognitive tasks were considered that were known to elicit non-obligatory brain responses, and the odd-ball task chosen. Three subjects listened to a series of two tones, one frequent and one rare, and counted the rare tones. During task performance, 40 to 46 channels of EEG activity were recorded from their scalps.

From the EEG data, average evoked potentials (aEP) were calculated for the frequent and rare conditions. From these a difference response was calculated. All three of these EPs were plotted as equipotential maps over a schematic of a head for topographic display and the major distribution features discussed. These aEPs and maps matched those previously reported in the literature.

From estimates of the spatial electrical power over the head, four peak components were selected for analysis by equivalent source modeling (ESM). These were designated the FP40, FP100, FP200, and FP350, where FP stands for field power. ESM demonstrated that one centrally located point dipole or two bilaterally symmetric dipoles could model the empirical data quite well. These results were discussed in relation to other topographic studies, as well as studies of intracranial recordings, lesions, and animal models. The source locations found were consistent with auditory cortical locations for the obligatory sensory peaks (FP40, FP100, FP200) and with brainstem locations as the source of the FP350 cognitive event-related peak.