VISUAL MOTION PERCEPTION IN LOW VISION

点击翻译按钮获取中文摘要

The long-term goal of this research program is to understand the relationship between vision loss and orientation/mobility (O/M). Because perception of the spatio-temporal properties of the environment is essential for successful O/M, it is important to study motion perception in people with vision loss. Anomalous visual motion perception has recently been observed in people with glaucoma and retinitis pigmentosa (RP). This could have serious ramifications for the use of motion information to guide mobility and control posture. The main goals of the proposed research are to characterize and understand visual motion processing in the central visual field of people with glaucoma and RP and to understand its role in O/M. Motion processing will be characterized by obtaining motion thresholds, investigating their dependence on stimulus parameters, and examining variation with disease progression. Bandpass- filtered spatial noise will be used to investigate the motion system in the central visual field of subjects with glaucoma, RP, and normal vision. Motion perception will be evaluated by obtaining thresholds of displacement, direction discrimination, and heading-direction discrimination as a function of spatial frequency, luminance, and speed. O/M performance will be characterized by obtaining measures of mobility, e.g. time to complete a predefined course and the number of contacts with obstacles, postural-sway, and responses to a questionnaire about arena of travel--life space. The role of motion processing in O/M will be assessed by comparing measures of O/M with motion thresholds obtained in the same subjects and determining the manner in which the measures vary as a function of stimulus and disease parameters. To better understand the mechanisms underlying motion processing in glaucoma and RP, we will use current motion models to computer simulate motion thresholds. Many motion models have as their basic components a spatial sampling array, spatial filters, and temporal filters. These parameters will be manipulated in the models to simulate expected outcomes. In addition, spatial and temporal CSFs will be measured, and sampling density will be estimated from frequency-of-seeing curves. These parameters will be used in the models to derive outcomes to compare to actual data. Given a better understanding of motion processing in glaucoma and RP, image-processing algorithms will be developed to attempt to compensate for anomalous motion processing and may lead to development of enhancement techniques for rehabilitation.