This project studies the dynamics of the VOR as an adaptive control system during the compensation periods of selected patients and the normal subjects undergoing high--fequency testing. The gain and phase of the horizontal and vertical vestibulo--ocular reflex (VOR) were determined at higher frequencies (2--6 Hz) using a protocal based on active head movements called the Vestibular Autorotation Test (VAT). Progress in this project focused on three areas: 1) Nonparametric gains and phases of patients with balance disorders, used for graphical comparison with normative data, 2) Parametric modeling using Auto--Regressive--Moving--Average (ARMA) methods, and 3) Block--structured modeling of the high--frequency horizontal and vertical VORs. Characteristics describing the gain and phase (vestibulograms), of the horizontal and vertical VOR were used to provide clinicians with patterns useful for determining diagnoses and to monitor changes in the VORs of patients with vestibular disorders undergoing treatment, i.e., ``balance retraining''. This treatment is based on using the brain's natural adaptive capabilities to respond to specific exercise therapy to alleviate symptoms of dysequilibrium. Parametric modeling of the VOR was done through ARMA modeling of patient data. Specific patient populations (e.g., Meniere's disease, gentamicin ototoxicity) were targeted to determine the ranges for parameters of a specific disease, and the parametric changes during therapeutic treatments which activated the adaptive control reduction of retinal--slip error signals in the VOR. Results showed significant differences from normal control groups. Results from specific disease categories are described below in collaborative studies. A new method was developed which provides quantitative predictions of future VOR parametric changes during clinical therapeutic treatment.
Showing the most recent 10 out of 84 publications
