| Dysfunction of central nervous system (CNS) caused by degenerative brain diseases, alcohol intoxication, chemical dependency, or trauma, is manifested in the performance of oculomotor system (OS). This study focuses on developing and evaluating a method that is sensitive to the functional changes in the OS. The method uses nonlinear analysis techniques to investigate the human OS, under normal and after alcohol treatment conditions. Monocular infrared oculogram (IROG) responses of 18 subjects, ages 21-30 years are recorded for pre- and post-alcohol treatment conditions, with a 6-frequency sum-of-sinusoids stimulus. The IROG responses are analyzed in frequency domain as well as in time domain. Frequency domain parameters such as spectral energy (power), peak frequencies, mean peak frequency, bandwidth and strength indices are calculated and compared between pre- and post-alcohol responses. Polynomial regression models are developed for the IROG responses of all subjects. The objectives of this study are to: (1) evaluate any OS dysfunction which may occur after ingestion of alcohol, using the spectral analysis of the OS responses, (2) study the frequency domain characteristics of the OS responses for pre- and post-alcohol treatments, (3) evaluate power strength indices of nonlinear frequencies, and (4) develop 4th degree polynomial regression models for the responses of all subjects.;From the statistical tests of frequency domain parameters--spectral power, peak frequencies, mean peak frequency, bandwidth, and strength indices--it is found that certain comparisons of primary and nonlinear frequencies are significantly different between pre- and post-alcohol treatment. This implies that some parameters show sensitivity to alcohol treatment, although a pattern of sensitivity in relation to blood alcohol concentration (BAC) of the subjects is not found. The combined spectral power of primary and nonlinear frequencies has an overall mean of 95.8%. The frequency response characteristics obtained from the data indicate that the responses have a low-pass filter type characteristic with dominant linear frequency components in the responses, which is also confirmed by the large F-ratio values (of 1st order coefficient) obtained from the regression analysis. The squared correlation coefficients of the left eye were significantly different between the pre- and post-alcohol responses. |
