The effect of retinal size and peripheral field restriction on the estimation of the speed of an automobile in a video scene was investigated. Sixteen participants of both genders with no driving experience estimated the speed of a car, while watching a video film recorded from the inside of the car. The video scene was recorded on a closed track, in absence of traffic. Two values of screen size and two values of distance to the screen were varied systematically. Two of the four combinations of screen size and distance corresponded to an identical retinal size. There were two conditions of peripheral visual information, which consisted of full screen or lateral restriction on both sides of the screen. There were eleven values of speed, which varied between 40km/h and 140km/h. All the subjects were tested under all the experimental conditions. Two values of speed were used as anchors to perform the estimation task. As general results, neither retinal size, nor peripheral field restriction showed any influence on speed estimation. The importance of the estimation method employed in order to limit the influence of these two variables (retinal size and peripheral field restriction) is discussed. The estimation pattern was similar in the four experimental conditions, with an underestimation trend for the highest speeds. The correlation analysis indicates that speed estimation is a highly stable ability of the subjects. (A)
Abstract