The field topography of evoked potentials is the dependent of the responses on the location of a small stimulus in the visual field. Field topographies are derived with a method of simultaneous stimulation of a large number of locations and extraction of the local responses from a single response signal. The techniques for such studies have been developed and tested. Goal 1: Determination of the field topography of the luminance and pattern ERGs and their components. Correlation of these topographies with densities of retinal receptors and ganglion cells. Inter-subject differences will be studied to establish a baseline for a clinical evaluation. Goal 2: Development of an objective test of local retinal function for the purpose of screening, diagnosis and monitoring of patients. The field plots will be derived from patients with specific etiologies (glaucoma, ocular hypertension, maculopathy). The plots will be compared with fundus images to establish the exact location of retinal areas with abnormal ERG responses. ERG fields will be compared with psychophysical fields to assess their sensitivity in the detection of local pathological changes. Goal 3: Identification and characterization of ERG components from different retinal layers. This will be achieved with different modes of pattern stimulation and techniques of nonlinear systems analysis. Goal 4: Identification of VEP components from different visual areas of cortex. Functional characterization of their sources on the basis of nonlinear systems analysis. The field topography of the VEP is very complex due to the convoluted cortical anatomy. For each stimulus location, the relative contributions to the response from various cortical sources are different. A principal component analysis (SVD) will be applied to the responses at locations of comparable eccentricity in the visual field, to determine the subspace spanned by the components from different sources. The kernel structure of the nonlinear responses will be used to identify and characterize the components within this subspace. The invariance of the components between subjects will be tested to confirm the decomposition.
